Diabetic retinopathy is the leading cause of blindness in adults living in developed countries.1 Almost all patients with type 1 diabetes mellitus (DM) and more than 60% of patients with type 2 DM will develop some degree of retinopathy after a 20-year history of diabetes.2 It has also been well established that DM increases the risk of cardiovascular disease.3 Cheng et al4 found that the prevalence of diabetic retinopathy associated with one, two, three, or four cardiometabolic risk factors was 16.0%, 17.6%, 21.3%, or 25.1%, respectively (P=0.001). This implies a relationship between systemic health conditions and diabetic retinopathy. In order to help identify and to prevent progression of this ophthalmic complication of diabetes, a better understanding of its association with systemic risk factors is necessary.

A PubMed literature search was conducted up to May 2016 using the following key words: “diabetic retinopathy”, “prevalence”, “epidemiology”, “systemic associations”, “risk factors”, “diabetic control”, “HbA1c”, “blood glucose”, “hypertension”, “hyperlipidaemia”, “cholesterol”, “obesity”, “smoking”, “myopia”, and “genetics”. Articles reporting systemic associations of diabetic retinopathy were selected and analysed. A search through the references of the retrieved articles was also performed. Only articles published in English were reviewed. During selection of the articles, prospective studies had a higher ranking than retrospective ones.

Approximately a quarter to one third of adults living with diabetes are reported to have diabetic retinopathy. In a recent systematic literature review, Yau et al5 estimated that the overall global prevalence of diabetic retinopathy was 34.6%. The Multi-Ethnic Study of Atherosclerosis (MESA) reported a prevalence of 33.2% in adults within the US population.6 A more recent study of the US population reports a lower prevalence of diabetic retinopathy of 28.5%.7

Within the Caucasian population in the US, the prevalence of diabetic retinopathy ranges from 24.8% to 26.4%.6 7 This is comparable with studies in western Europe with largely Caucasian populations, such as in the Gutenberg Health Study in Germany with a prevalence of 21.7%8 and Tromsø Eye Study in Norway with a prevalence of 26.8%.9

Multiple studies have reported a higher prevalence of diabetic retinopathy in black and Hispanic populations. Studies performed in the US reported a prevalence of 36.7%6 and 38.8%7 in African American populations. Additionally, the Los Angeles Latino Eye Study (LALES) reported that 46% of Hispanic Americans with type 2 DM had diabetic retinopathy.10 The MESA study6 and a more recent study by Zhang et al7 reported a lower prevalence of 37.4% and 34.0%, respectively in the Hispanic American population.

There is considerable variation in prevalence rates of diabetic retinopathy within the Chinese population. The prevalence in Hong Kong has been reported between 28.4% and 39.2% in different studies.11 12 13 A more recent study by Lian et al14 on a local screening programme for diabetic retinopathy reported a prevalence rate of 39.0% (95% confidence interval [CI], 38.8%-39.2%) in Hong Kong. Chen et al15 reported a prevalence of 35.0% in Taiwanese diabetic patients. Similarly, the comprehensive Beijing Eye Study covering both urban and rural Chinese populations reported a prevalence of 37%,16 in comparison with the Handan Eye Study focusing on rural China that reported a higher prevalence of 43.1%.17 The MESA study included data on Chinese Americans and reported a prevalence rate of 25.7%, comparable with their Caucasian counterparts.6

Differences in prevalence rates of diabetic retinopathy between populations may be due to both genetic factors and access to health care. Moreover, ethnicity is complex and multifactorial; alone it may not fully explain the differences in prevalence rates of diabetic retinopathy reported. Urbanisation and socio-economic status may also play a role in the prevalence of diabetic retinopathy.18 Additionally, discrepancies between studies of diabetic retinopathy prevalence may depend on several factors including study methodology. Some studies may have used different cut-off values for glycated haemoglobin (HbA1c), oral glucose tolerance test, and spot glucose to define a diagnosis of DM, thereby increasing the heterogeneity in the overall study population. The method of diabetic retinopathy screening and reporting logistics also varied among studies. For studies using retinal photographs as screening tools, the differences in the number of fundus photographs taken and whether pupil dilatation was performed before retinal assessment may have influenced the number of cases of diabetic retinopathy identified.

The Early Treatment Diabetic Retinopathy Study (ETDRS) Diabetic Retinopathy Grading Scale (Modified Airlie House) and International Clinical Diabetic Retinopathy (ICDR) Severity Scale are the two major and most commonly used classification systems for diabetic retinopathy. The ETDRS Grading Scale is more commonly used in research contexts whereas the ICDR Severity Scale is more commonly used in the clinical context. Additionally, the United Kingdom National Screening Committee (UK NSC) diabetic retinopathy grading system is notable for its widespread use in digital fundus photo screening programmes worldwide.

In general, diabetic retinopathy can be classified as non-proliferative or proliferative. Non-proliferative can then be further classified by severity ranging from mild to moderate and severe. Non-proliferative diabetic retinopathy (NPDR) is characterised by the presence of microaneurysms, hard exudates, cotton-wool spots, and/or retinal haemorrhages. Pre-proliferative diabetic retinopathy changes include vasculopathies such as intraretinal microvascular abnormality (IRMA) whereas proliferative diabetic retinopathy is defined by the presence of neovascularisation or vitreous haemorrhage or preretinal haemorrhage.

The ETDRS Diabetic Retinopathy Grading Scale assigns a retinopathy severity score from level 10 to 85. If no abnormality is found, ETDRS level 10 is assigned. If only microaneurysms are present, ETDRS level 20 (very mild NPDR) is assigned. The ETDRS level 35 is equivalent to mild NPDR, and is characterised by the presence of hard exudates, cotton-wool spots, and/or mild retinal haemorrhages. Standard retinal photographs of retinal haemorrhages and IRMA are used to define moderate and severe NPDR in the ETDRS grading. The number of quadrants in which those signs are present is also taken into account. In contrast, a simple 4-2-1 rule is used in the ICDR Severity Scale: mild NPDR is defined as the presence of microaneurysms only. Severe NPDR is diagnosed when there are ≥20 diffuse intraretinal haemorrhages and/or microaneurysms in all four quadrants, venous beading in at least two quadrants, or intraretinal microvascular abnormalities in at least one quadrant. Moderate NPDR on the other hand has severity falling below the 4-2-1 rule but more than merely microaneurysms. In both grading scales, proliferative diabetic retinopathy is defined as the presence of neovascularisation. High-risk and advanced proliferative diabetic retinopathy (ETDRS level, 71-85) refers to the presence of neovascularisation complications including preretinal haemorrhage, vitreous haemorrhage, or retinal detachment.19

Under the UK NSC diabetic retinopathy screening system, fundus photos are given a letter and numerical grading: R0 is assigned if no abnormality is found; R1 is given in early diabetic retinopathy with evidence of microaneurysms, retinal haemorrhage, or other features of diabetic retinopathy; R2 and R3 ratings are assigned in pre-proliferative and proliferative diabetic retinopathy, respectively. M1 is designated in cases with maculopathy, the presence of which is used to predict the presence of clinically significant macular oedema that requires treatment. P1 is assigned if the fundus photograph shows evidence of previous panretinal laser photocoagulation. Accordingly M0 and P0 are given respectively if the above characteristics are absent. Finally, a grading of U is given if an image is regarded as ungradable,20 common reasons include dense cataract and corneal scars.

Glycated haemoglobin is a commonly used marker for monitoring glycaemic control. Multiple studies have consistently shown HbA1c to be an independent risk factor for diabetic retinopathy.6 7 10 21 22 A higher HbA1c is associated with both increased incidence as well as progression of diabetic retinopathy.21 The LALES study found a 22% increase in prevalence of diabetic retinopathy with 1% increase in HbA1c. Data suggested a plateau of the curve at HbA1c of ≥11%, however.10 Elevated HbA1c reflects poorly controlled diabetes, which is one of the major causes of complications in DM including diabetic retinopathy. Nonetheless, the United Kingdom Prospective Diabetes Study (UKPDS) demonstrated that even a good HbA1c level of 7.0% had an absolute risk of 7.9 per 1000 patient-years for retinal laser photocoagulation.10 23 This emphasises the importance of optimal glycaemic control in diabetic patients in order to prevent diabetic retinopathy. In contrast, this association was not seen in young-onset type 1 DM,24 a difference that may be attributed to the role of hyperglycaemic memory.25

The ACCORD study26 compared the effects of intensive glycaemic treatment with target HbA1c of <6.0% compared with standard treatment with target HbA1c of 7.0% to 7.9%. The authors found a decreased rate of progression of diabetic retinopathy in the intensive treatment group, 7.3%, versus 10.4% in the standard therapy group (adjusted odds ratio [OR]=0.67; 95% CI, 0.51-0.87; P=0.003).26 Despite the perceived benefit of reduced progression of diabetic retinopathy, intensive glycaemic control is not without risk. The ACCORD study found an increased risk of hypoglycaemia requiring medical assistance (10.5% vs 3.5%; P=0.001), weight gain of >10 kg (27.8% vs 14.1%; P=0.001), and all-cause mortality (hazard ratio=1.22; 95% CI, 1.01-1.46; P=0.04).27 Similar findings have also been demonstrated in the Diabetes Control and Complications Trial (DCCT).28 The authors reported that intensive diabetes control therapy reduced the adjusted mean risk for development of retinopathy by 76%, but at a cost of a 33% increased risk of weight gain and a 3-times higher risk of severe hypoglycaemia.28 Thus, one must balance the benefit of reducing the risk of diabetic retinopathy and the risk of tight glycaemic control when treating DM. As demonstrated in this review, however, multiple factors play a role in the development and progression of diabetic retinopathy, hence tight blood glucose control alone may not entirely prevent the development of diabetic retinopathy. Furthermore, Mohamed et al29 have recommended that in patients with diabetic retinopathy, a HbA1c level of 7% is ideal in reducing progression of and new development of diabetic retinopathy.

Studies have also found a significantly higher fasting plasma glucose in subjects with diabetic retinopathy.6 16 Xie et al16 found that patients with diabetic retinopathy had a mean (± standard deviation) fasting plasma glucose level of 8.88 ± 4.56 mmol/L compared with 7.70 ± 2.80 mmol/L in those without diabetic retinopathy. This association may also be due to the effects of hyperglycaemia causing retinopathy.

An unmodifiable risk factor, prolonged duration of diabetes, has been consistently demonstrated to be a risk factor for diabetic retinopathy.6 8 16 21 24 30 In fact, one study reported that patients with diabetic retinopathy had longer duration of diabetes, double than those without retinopathy (25 ± 10 vs 12 ± 8 years; P<0.0001).30 This was corroborated by Zhang et al7 in a large-scale study which found that patients with diabetic retinopathy had a longer duration of diabetes (15.0 ± 1.6 years vs 7.3 ± 0.8 years; P<0.001). Furthermore, Wong et al31 reported the OR of diabetic retinopathy increased by 1.07 ± 0.2 per year of duration of disease. The LALES study found that each year of increased history of diabetes was associated with an 8% increased risk of having diabetic retinopathy.10 This association can be explained by a prolonged exposure to the hyperglycaemic state that may increase the risk of vascular injury, leading to diabetic retinopathy and other complications.

Patients with diabetic retinopathy are also more likely to require medication such as oral hypoglycaemic agents or insulin to control their diabetes. In other words, known and treated diabetes is a predictor of diabetic retinopathy compared with unknown and untreated diabetes.32 A large-scale study performed in the Chinese population found that 90% of diabetic patients without retinopathy were either not on treatment, diet control, or oral hypoglycaemic agents.16 In contrast, almost 80% of patients with diabetic retinopathy required oral hypoglycaemic agents or insulin injections for diabetic control.16 Diabetic patients with diabetic retinopathy were statistically significantly more likely to use insulin (47.4 ± 8.3% vs 26.7 ± 4.8%; OR=3.23).7 One study found the prevalence of diabetic retinopathy to be 70% in patients with type 2 DM using insulin, compared with only 39% in those not receiving insulin treatment.33 These findings are consistent with several other studies.10 34 This relationship between insulin use and diabetic retinopathy may be explained by the severity and level of blood glucose control in patients. In other words, patients who did not require medication are likely those with borderline diabetes or relatively well controlled blood glucose profile, and thus have less risk of developing diabetic retinopathy.

The associations between diabetic risk factors and diabetic retinopathy are summarised in Table 1.6 7 8 16 21 24 29 30

Hypertension has been consistently demonstrated to have a positive association with the development of diabetic retinopathy.4 6 7 8 30 The LALES study found an OR of 1.26 (P=0.002) for every 20 mm Hg increase in blood pressure.10 The Hoorn study estimated that patients with hypertension had more than double the risk of developing retinopathy after 10 years when compared with diabetic patients with normal blood pressure.22 Stratton et al35 found that the incidence of developing new retinopathy increased from 17% to 32% when comparing the lowest tertile with the top third mean blood pressure in patients with diabetic retinopathy (P<0.0001). This gradient was less pronounced with 26% and 36% of progression in retinopathy when the bottom third was compared with the top tertile of blood pressure, respectively (P=0.005).35 This marked association between hypertension and diabetic retinopathy may be explained by the clinical finding that hypertension and diabetes frequently co-exist. Hypertension may cause morphological changes in the retinal vessels that are similar to those seen in mild-to-moderate NPDR such as hard exudates, cotton-wool spots, and retinal haemorrhages.34

The landmark UKPDS 69 study has outlined the importance of blood pressure control in patients with diabetic retinopathy.21 The authors demonstrated that tight control of blood pressure with a target level of 150/85 mm Hg, rather than loose control of less than 180/105 mm Hg, statistically significantly decreased the development of microaneurysms (relative risk [RR]=0.66; P<0.001), hard exudates (RR=0.53; P<0.001), and cotton-wool spots (RR=0.53; P<0.001). These effects were evident by 4.5 years of follow-up and persisted for up to 7.5 years. Furthermore, there were no detectable differences between blood pressure control by atenolol or captopril therapy, nor in primary or secondary prevention groups. Additionally, the loose blood pressure control group had an absolute risk of 4.1 per 1000 patient-years for blindness in one eye due to all causes (P=0.046; RR=0.76; 99% CI, 0.29-1.99), in comparison with an absolute risk of 3.1 per 1000 patient-years risk of blindness in the tight blood pressure control group.21

The importance of tight blood pressure control could not be clearly demonstrated in both the Appropriate Blood Pressure Control in Diabetes (ABCD) Trial36 or the ACCORD study.26 In the ABCD Trial, subjects were randomised to intensive control group (diastolic blood pressure of 75 mm Hg) or moderate control group (diastolic blood pressure of 80-89 mm Hg).36 The authors found no statistically significant difference in the progression of diabetic retinopathy over a 5-year follow-up period. Discrepancies in findings between the studies may be partially due to the difference in blood pressure targets. The authors of the ABCD Trial found that both groups had poor glycaemic control despite optimal blood pressure control and this may have accounted for the progression of diabetic retinopathy.36 This may further suggest the importance of glycaemic control in diabetic retinopathy. Likewise, the ACCORD study found no statistically significant relationship between intensive blood pressure therapy (median systolic blood pressure, 117 mm Hg) and standard therapy (median systolic blood pressure, 133 mm Hg) [10.4% vs 8.8%; adjusted OR=1.23; 95% CI, 0.84-1.79; P=0.29].26 The authors hypothesised that these findings may be attributable to the small difference between blood pressure in the two groups.

Obesity is another risk factor commonly associated with cardiovascular disease. It can be defined by waist-hip ratio, waist circumference, and body mass index (BMI). Both greater waist-hip ratio and waist circumference are positively associated with diabetic retinopathy.6 22 24 The OR of diabetic retinopathy is 1.28 per 5-cm increase in waist circumference (OR=1.28; 95% CI, 1.05-1.56; P=0.014).24

The association between BMI and diabetic retinopathy has been variable among studies. A study looking specifically at patients with type 1 DM found that obesity with a BMI of >30 kg/m² was the predominant risk factor for diabetic retinopathy, even when controlling for other risk factors such as HbA1c and the use of cardioprotective drugs.37 In another study, obesity was associated with increased prevalence of retinopathy and patients with retinopathy were more likely to be obese, although this association was not sustained after adjusting for confounding factors such as blood pressure.30 Yet among patients with retinopathy, higher BMI was noted to be positively associated with more severe retinopathy and vision-threatening retinopathy.8

Some studies have reported no increased risk between elevated BMI and retinopathy.38 39 On the contrary, some studies have even reported an inverse relationship between BMI and diabetic retinopathy. The Wisconsin Epidemiologic Study of Diabetic Retinopathy found that underweight patients (BMI <20 kg/m²) with diabetes had a higher incidence of retinopathy compared with obese patients (RR=1.99; 95% CI, 1.21-3.26 vs RR=1.27; 95% CI, 1.00-1.61).40 In fact, 100% of the underweight patients in this study developed retinopathy by 10 years. This observation held true for the progression of retinopathy as well. The authors observed that the underweight patients had a longer duration of diabetes and were also more likely to be taking insulin than the obese subjects. Thus, they hypothesised that the underweight patients might have had poorer overall glycaemic control and hence were in a more ‘severe’ phase of diabetes when compared with their obese counterparts.40 This hypothesis also corroborates the findings of the DCCT study that tight glycaemic control increases the risk of being overweight.28 Although not statistically significant, Wong et al31 reported that a lower BMI was associated with diabetic retinopathy.

Male sex is an independent risk factor for diabetic retinopathy. A large-scale study performed in the United States revealed that in diabetic patients over the age of 40 years, 38% ± 5.5% of men compared with 27.1% ± 4.7% of women had diabetic retinopathy (OR=2.07; 95% CI, 1.39-3.10).7 While the LALES study showed no statistically significant difference in the incidence of diabetic retinopathy between different sexes, their stepwise multivariate model demonstrated that men had a 50% higher risk of having any diabetic retinopathy when compared with females (OR=1.50; P=0.006).10 This finding was echoed by the UKPDS 50 study that also found no difference in incidence between the two sexes (P=0.67), but a multivariate model showed that women had a lower RR of progression of diabetic retinopathy (RR=0.54; CI, 0.37-0.80; P=0.0016).35 Despite the above evidence, several other studies have found no statistically significant associations between diabetic retinopathy and sex.22 30

Studies have found varying relationships between elevated cholesterol and diabetic retinopathy. While Yau et al5 reported that elevated total serum cholesterol was associated with a higher prevalence of diabetic macular oedema and vision-threatening diabetic retinopathy, other studies have been unable to reproduce similar results. Tomi&cacute; et al34 demonstrated no statistically significant difference in cholesterol level between patients with different severity of diabetic retinopathy. The Hoorn study also found no relationship between total cholesterol level and incidence of diabetic retinopathy,22 but demonstrated that elevated serum lipid level is associated with increased prevalence of hard exudates characterising NPDR.41 This association of elevated serum lipid levels and hard exudates was observed in other studies as well.42 43 44 Patients with type 1 DM with diabetic retinopathy have been shown to have statistically significantly higher total cholesterol level than those without diabetic retinopathy (199 ± 35 mg/dL vs 188 ± 36 mg/dL; P=0.001), although after logistic regression analysis this was not shown to be an independent risk factor for diabetic retinopathy.30 On the contrary, Wong et al31 reported higher total cholesterol level to be protective of diabetic retinopathy (OR=0.73, per 1 mmol/L increase).

Studies showed inconsistent results in the relationship between serum triglycerides and diabetic retinopathy. The Hoorn study,22 De Block et al,30 and Tomi&cacute; et al34 have all demonstrated that serum triglyceride levels are not an independent risk factor for diabetic retinopathy. Likewise, in a large-scale study of 2535 patients with type 2 DM, retinopathy was not significantly associated with triglyceride or high-density lipoprotein (HDL) cholesterol levels after adjustment of confounding factors.45 In contrast, Cheng at al4 found that in overweight type 2 DM patients, elevated triacylglycerol levels were significantly associated with diabetic retinopathy (OR=1.29; 95% CI, 1.05-1.58; P<0.05).

Despite the conflicting data, there appears to be merit in treating hyperlipidaemia in patients with diabetic retinopathy. The ACCORD trial investigated the effects of intensive treatment of dyslipidaemia on the progression of diabetic retinopathy.26 Patients in the intensive treatment group were given 160 mg of fenofibrate daily plus simvastatin, while patients in the standard treatment group were given placebo with simvastatin. The authors found a significant improvement in HDL cholesterol (P=0.002) and significant decrease in triglyceride level (P<0.001) in the intensive treatment group. As expected, fall in low-density lipoprotein cholesterol level was comparable between the two groups (P=0.68) as both received simvastatin. With the significant improvement in lipid profile, the authors found that intensive therapy decreased the risk of progression of diabetic retinopathy compared with simvastatin alone, 6.5% vs 10.2% (adjusted OR=0.60; 95% CI, 0.42-0.87; P=0.006).26 A more recent meta-analysis by Das et al46 found no statistically significant improvement in severity of diabetic macular oedema or decrease in progression of hard exudates in patients receiving lipid-lowering drugs compared with placebo. The authors, however, remarked that studies included in the meta-analysis were of questionable quality, calling to the necessity for further high-quality research. Studies included in the meta-analysis by Das et al46 had small sample sizes, underpowered studies, and poor subject selection criteria since patients included had early-stage diabetic retinopathy and hence were at low risk of progression.

Both retinopathy and nephropathy are microvascular complications of diabetic retinopathy. Multiple studies have demonstrated the association between diabetic retinopathy and chronic kidney disease. In a study by Park et al47 in a Korean population, the authors defined chronic kidney disease as estimated glomerular filtration rate (eGFR) of <60 mL/min/1.73 m³. They reported that even after controlling for confounders, both chronic kidney disease (OR=2.34; 95% CI, 1.04-5.28) and proteinuria (OR=4.56; 95% CI, 1.51-13.77) were significantly associated with diabetic retinopathy.47 Additionally, Zhang et al48 found that in the Chinese population, lower eGFR was significantly associated with increasing severity of diabetic retinopathy in patients with diabetic retinopathy (mean eGFR, 93 mL/min/1.73 m³) compared with patients without (mean eGFR, 116 mL/min/1.73 m³; P<0.0001), independent of hypertension and diabetes duration. Diabetic retinopathy was also associated with microalbuminuria (P<0.0001) and higher albumin-to-creatinine ratio (6.4 vs 0.63 in patients with and without diabetic retinopathy, respectively). The authors suggested that in diabetic patients, chronic hyperglycaemia causes microvascular changes in both the glomerulus of the kidney and retina of the eye. Over time, these microvascular changes lead to narrowing and occlusion of the vascular lumina, and eventually cause inadequate perfusion of affected tissues leading to retinopathy and nephropathy.48 These findings were corroborated by Penno et al49 who reported that a high urine albumin-to-creatinine ratio of ≥300 mg/g was associated with diabetic retinopathy (OR=2.9; 95% CI, 2.1-4.0). Likewise, Rodríguez-Poncelas et al50 found that increasing urine albumin-to-creatinine ratio was significantly correlated with rising diabetic retinopathy prevalence, and this association was significant even at urine albumin-to-creatinine levels of ≥10 mg/g (OR=1.2; 95% CI, 1.1-1.4).

Smoking was shown to have neither a statistically significantly positive nor negative association with diabetic retinopathy. The Hoorn study demonstrated a non-significant trend for increased OR of diabetic retinopathy incidence in cigarette smokers and ex-smokers.22 A 25-year follow-up study showed a non-significant trend that current smokers were more likely to develop proliferative diabetic retinopathy than never smokers, yet was unable to establish a statistically significant association between smoking status or pack-years of smoking and proliferative diabetic retinopathy. The authors found that mild NPDR was more common among current smokers than former smokers (28.4% vs 13.0%; P=0.038) and may suggest that smoking is indeed related to early forms of diabetic retinopathy.51 On the contrary, the UKPDS 50 study demonstrated a protective effect of smoking—current smokers had a reduced incidence of retinopathy with a RR of 0.63 (95% CI, 0.4-0.82; P=0.0043) as well as reduced progression of retinopathy with a RR of 0.50 (95% CI, 0.36-0.71, P=0.0045) when compared with never smokers.35

Long-sightedness is prevalent among Asians and it has become a significant problem in this locality in terms of resources for glasses prescriptions, refractive surgeries, and the management of visual-threatening complications such as myopic macular degeneration, myopic tractional maculopathy, and choroidal neovascularisation. Although myopia in most cases is harmful to ocular health, it has long been observed that the prevalence of diabetic retinopathy is low in myopic patients. This protective effect of myopia against diabetic retinopathy has recently been proven by various meta-analysis,52 53 revealing the OR of diabetic retinopathy in myopic diabetic patients versus non-myopic diabetic patients to be 0.70 (95% CI, 0.58-0.85; P<0.001). On the contrary, axial length, the major cause of myopia, was found to be associated with diabetic retinopathy as well, in which each millimetre increase in axial length is associated with decreased risk of diabetic retinopathy (OR=0.75; 95% CI, 0.65-0.86; P<0.001).

The relationships between systemic risk factors and diabetic retinopathy are summarised in Table 2.4 5 6 7 8 10 19 21 22 24 28 30 34 38 39 40 42 43 44 47 48 49 50 51 52 53

While genetic polymorphism is not a modifiable risk factor, various genetic polymorphisms may cause a predisposition to the development and progression of diabetic retinopathy. Multiple studies have demonstrated the relationship between TCF7L2 and the development of type 2 diabetes54 55 56 as increased expression of this gene is associated with poor serum glucose control. Ciccacci et al57 found that patients with rs12255372 or rs7903146 variants of TCF7L2 were at higher risk of developing diabetic retinopathy. Within the Chinese population, patients with rs6585205, rs7903146, and rs11196218 had a weakly positive but not statistically significant association with development of diabetic retinopathy.58 A meta-analysis conducted by Ding et al59 found that the rs7903146 variant (T allele) of TCF7L2 was significantly associated with increased risk of development of diabetic retinopathy (OR=1.47; 95% CI, 1.19-1.81; P≤0.001 for TT vs CC when comparing genotype polymorphism TT, TC, and CC) especially within the Caucasian population. Due to rarity of the T allele within East Asian populations (estimated at 4.47%60 and 6.997%58 in two different studies), however, this may be not applicable to this locality.

A meta-analysis by Ma et al61 reported a positive association between Pro12Ala polymorphism of the peroxisome proliferator–activated receptor γ2 (PPARγ2) gene. The PPARγ2 gene plays a key role in multiple pathways including glucose metabolism, angiogenesis, and inflammation. The authors found that the Ala allele yielded a protective effect against diabetic retinopathy in patients with type 2 DM (OR=0.81; 95% CI, 0.68-0.98; P=0.03). This association was stronger in Caucasian subgroups compared with Asian subgroups, possibly due to differences in allele frequencies and study design.61

A recent study by Peng et al62 demonstrated that the C-reactive protein (CRP) variant rs2808629 is statistically significantly associated with increased risk of diabetic retinopathy (OR=1.296; 95% CI, 1.076-1.561; P=0.006 for G allele) in Chinese patients with type 2 DM. Even after adjusting for confounding factors associated with diabetic retinopathy, this variant of CRP remained an independent genetic risk factor for development of diabetic retinopathy.62 Patients with rs2808629 have been demonstrated to have higher levels of serum CRP.63 Yet, there have been no conclusive findings of the relationship between high CRP levels and the development of diabetic retinopathy as studies have demonstrated contradicting results.64 65 66 67 68 The exact mechanism of how rs2808629 causes increased risk of diabetic retinopathy warrants further research.

Another comprehensive study by Peng et al69 sought to establish the relationship between 40 single nucleotide polymorphisms and diabetic retinopathy in Chinese patients with type 2 DM. The authors found that rs17684886 in ZNRF1 (OR=0.812; P=0.0039) and rs599019 near COLEC12 (OR=0.835; P=0.0116) were associated with an increased risk of diabetic retinopathy. rs6427247 near SCYL1BP1 (OR=1.368; P=0.0333) and rs899036 near API5 (OR=0.340; P=0.0005) were associated with increased risk of severe diabetic retinopathy.69 This is consistent with other genome-wide association studies in Caucasian and Mexican-American patients.70 71

As the relationship between genetic polymorphisms and diabetic retinopathy is still a new and emerging field, some studies have demonstrated new associations between single nucleotide polymorphisms, but these studies have not yet been replicable. Further studies are thus warranted.

Diabetic retinopathy and its genetic risk factors are listed in Table 3.57 58 60 61 62 63 69 70 71

The importance of glycaemic control and duration of diabetes with diabetic retinopathy have been clearly established. Additionally, the modality of diabetic treatment may reflect the severity of diabetes and risk of developing diabetic retinopathy. Patients with DM should be encouraged to optimise their control of the disease in order to prevent the development and progression of diabetic retinopathy. Of the systemic risk factors studied, multiple studies clearly establish a positive association between hypertension and diabetic retinopathy. Studies have shown varying results for the association of diabetic retinopathy with obesity, male sex, hyperlipidaemia, and smoking. Additionally, declining renal function and microalbuminuria have been demonstrated to be associated with increased prevalence of diabetic retinopathy. In contrast, myopia is protective against development of diabetic retinopathy. Despite inconsistent findings for the association of systemic risk factors with diabetic retinopathy, it is still important for clinicians to encourage patients to optimise their body weight, lipid profile, and to abstain from smoking due to their associations with risk for cardiovascular disease as well as other complications of diabetes. New research has also demonstrated an increasing number of genetic polymorphisms associated with risk of type 2 DM and the development of diabetic retinopathy. Investigating the polymorphisms associated with diabetic retinopathy may help us better understand the developmental pathway of diabetic retinopathy and lead to new targeted therapy in treating diabetes and preventing diabetic retinopathy.

All authors have disclosed no conflicts of interest.

1. Cheung N, Mitchell P, Wong TY. Diabetic retinopathy. Lancet 2010;376:124-36. Crossref

2. Klein R, Klein BE, Moss SE, Cruickshanks KJ. The Wisconsin Epidemiologic Study of Diabetic Retinopathy: XVII. The 14-year incidence and progression of diabetic retinopathy and associated risk factors in type 1 diabetes. Ophthalmology 1998;105:1801-15. Crossref

3. Kannel WB, McGee DL. Diabetes and cardiovascular risk factors: the Framingham study. Circulation 1979;59:8-13. Crossref

4. Cheng Y, Zhang H, Chen R, et al. Cardiometabolic risk profiles associated with chronic complications in overweight and obese type 2 diabetes patients in South China. PLoS One 2014;9:e101289. Crossref

5. Yau JW, Rogers SL, Kawasaki R, et al. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care 2012;35:556-64. Crossref

6. Wong TY, Klein R, Islam FM, et al. Diabetic retinopathy in a multi-ethnic cohort in the United States. Am J Ophthalmol 2006;141:446-55. Crossref

7. Zhang X, Saaddine JB, Chou CF, et al. Prevalence of diabetic retinopathy in the United States, 2005-2008. JAMA 2010;304:649-56. Crossref

8. Raum P, Lamparter J, Ponto KA, et al. Prevalence and cardiovascular associations of diabetic retinopathy and maculopathy: results from the Gutenberg Health Study. PLoS One 2015;10:e0127188. Crossref

9. Bertelsen G, Peto T, Lindekleiv H, et al. Sex differences in risk factors for retinopathy in non-diabetic men and women: the Tromsø Eye Study. Acta Ophthalmol 2014;92:316-22. Crossref

10. Varma R, Macias GL, Torres M, et al. Biologic risk factors associated with diabetic retinopathy: the Los Angeles Latino Eye Study. Ophthalmology 2007;114:1332-40. Crossref

11. Tam TK, Lau CM, Tsang LC, Ng KK, Ho KS, Lai TC. Epidemiological study of diabetic retinopathy in a primary care setting in Hong Kong. Hong Kong Med J 2005;11:438-44.

12. Tam VH, Lam EP, Chu BC, Tse KK, Fung LM. Incidence and progression of diabetic retinopathy in Hong Kong Chinese with type 2 diabetes mellitus. J Diabetes Complications 2009;23:185-93. Crossref

13. Wang WQ, Ip TP, Lam KS. Changing prevalence of retinopathy in newly diagnosed non-insulin dependent diabetes mellitus patients in Hong Kong. Diabetes Res Clin Pract 1998;39:185-91. Crossref

14. Lian JX, Gangwani RA, McGhee SM, et al. Systematic screening for diabetic retinopathy (DR) in Hong Kong: prevalence of DR and visual impairment among diabetic population. Br J Ophthalmol 2016;100:151-5. Crossref

15. Chen MS, Kao CS, Chang CJ, et al. Prevalence and risk factors of diabetic retinopathy among noninsulin-dependent diabetic subjects. Am J Ophthalmol 1992;114:723-30. Crossref

16. Xie XW, Xu L, Wang YX, Jonas JB. Prevalence and associated factors of diabetic retinopathy. The Beijing Eye Study 2006. Graefes Arch Clin Exp Ophthalmol 2008;246:1519-26. Crossref

17. Wang FH, Liang YB, Zhang F, et al. Prevalence of diabetic retinopathy in rural China: the Handan Eye Study. Ophthalmology 2009;116:461-7. Crossref

18. Ramachandran A, Snehalatha C, Vijay V, King H. Impact of poverty on the prevalence of diabetes and its complications in urban southern India. Diabet Med 2002;19:130-5. Crossref

19. Grading diabetic retinopathy from stereoscopic color fundus photographs—an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology 1991;98(5 Suppl):786S-806S. Crossref

20. Core NDESP team. Diabetic eye screening feature based grading forms. NHS Diabetic Eye Screening Programme; 2012.

21. Matthews DR, Stratton IM, Aldington SJ, Holman RR, Kohner EM, UK Prospective Diabetes Study Group. Risks of progression of retinopathy and vision loss related to tight blood pressure control in type 2 diabetes mellitus: UKPDS 69. Arch Ophthalmol 2004;122:1631-40. Crossref

22. van Leiden HA, Dekker JM, Moll AC, et al. Risk factors for incident retinopathy in a diabetic and nondiabetic population: the Hoorn study. Arch Ophthalmol 2003;121:245-51. Crossref

23. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998;352:837-53. Crossref

24. Rajalakshmi R, Amutha A, Ranjani H, et al. Prevalence and risk factors for diabetic retinopathy in Asian Indians with young onset type 1 and type 2 diabetes. J Diabetes Complications 2014;28:291-7. Crossref

25. Zhang L, Chen B, Tang L. Metabolic memory: mechanisms and implications for diabetic retinopathy. Diabetes Res Clin Pract 2012;96:286-93. Crossref

26. ACCORD Study Group, ACCORD Eye Study Group, Chew EY, et al. Effects of medical therapies on retinopathy progression in type 2 diabetes. N Engl J Med 2010;363:233-44. Crossref

27. Action to Control Cardiovascular Risk in Diabetes Study Group, Gerstein HC, Miller ME, et al. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med 2008;358:2545-59. Crossref

28. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med 1993;329:977-86. Crossref

29. Mohamed Q, Gillies MC, Wong TY. Management of diabetic retinopathy: a systematic review. JAMA 2007;298:902-16. Crossref

30. De Block CE, De Leeuw IH, Van Gaal LF. Impact of overweight on chronic microvascular complications in type 1 diabetic patients. Diabetes Care 2005;28:1649-55. Crossref

31. Wong TY, Cheung N, Tay WT, et al. Prevalence and risk factors for diabetic retinopathy: the Singapore Malay Eye Study. Ophthalmology 2008;115:1869-75. Crossref

32. Lim MC, Lee SY, Cheng BC, et al. Diabetic retinopathy in diabetics referred to a tertiary centre from a nationwide screening programme. Ann Acad Med Singapore 2008;37:753-9.

33. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wisconsin epidemiologic study of diabetic retinopathy. III. Prevalence and risk of diabetic retinopathy when age at diagnosis is 30 or more years. Arch Ophthalmol 1984;102:527-32. Crossref

34. Tomi&cacute; M, Ljubi&cacute; S, Kaštelan S, Gverovi&cacute; Antunica A, Jazbec A, Polji&ccaron;anin T. Inflammation, haemostatic disturbance, and obesity: possible link to pathogenesis of diabetic retinopathy in type 2 diabetes. Mediators Inflamm 2013;2013:818671. Crossref

35. Stratton IM, Kohner EM, Aldington SJ, et al. UKPDS 50: risk factors for incidence and progression of retinopathy in type II diabetes over 6 years from diagnosis. Diabetologia 2001;44:156-63. Crossref

36. Estacio RO, Jeffers BW, Gifford N, Schrier RW. Effect of blood pressure control on diabetic microvascular complications in patients with hypertension and type 2 diabetes. Diabetes Care 2000;23 Suppl 2:B54-64.

37. Price SA, Gorelik A, Fourlanos S, Colman PG, Wentworth JM. Obesity is associated with retinopathy and macrovascular disease in type 1 diabetes. Obes Res Clin Pract 2014;8:e178-82. Crossref

38. Nelson RG, Wolfe JA, Horton MB, Pettitt DJ, Bennett PH, Knowler WC. Proliferative retinopathy in NIDDM. Incidence and risk factors in Pima Indians. Diabetes 1989;38:435-40. Crossref

39. Lee ET, Lee VS, Lu M, Russell D. Development of proliferative retinopathy in NIDDM. A follow-up study of American Indians in Oklahoma. Diabetes 1992;41:359-67. Crossref

40. Klein R, Klein BE, Moss SE. Is obesity related to microvascular and macrovascular complications in diabetes? The Wisconsin Epidemiologic Study of Diabetic Retinopathy. Arch Intern Med 1997;157:650-6. Crossref

41. van Leiden HA, Dekker JM, Moll AC, et al. Blood pressure, lipids, and obesity are associated with retinopathy: the Hoorn study. Diabetes Care 2002;25:1320-5. Crossref

42. Klein BE, Moss SE, Klein R, Surawicz TS. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. XIII. Relationship of serum cholesterol to retinopathy and hard exudate. Ophthalmology 1991;98:1261-5. Crossref

43. Chew EY, Klein ML, Ferris FL 3rd, et al. Association of elevated serum lipid levels with retinal hard exudate in diabetic retinopathy. Early Treatment Diabetic Retinopathy Study (ETDRS) Report 22. Arch Ophthalmol 1996;114:1079-84. Crossref

44. Chaturvedi N, Sjoelie AK, Porta M, et al. Markers of insulin resistance are strong risk factors for retinopathy incidence in type 1 diabetes. Diabetes Care 2001;24:284-9. Crossref

45. Sacks FM, Hermans MP, Fioretto P, et al. Association between plasma triglycerides and high-density lipoprotein cholesterol and microvascular kidney disease and retinopathy in type 2 diabetes mellitus: a global case-control study in 13 countries. Circulation 2014;129:999-1008. Crossref

46. Das R, Kerr R, Chakravarthy U, Hogg RE. Dyslipidemia and diabetic macular edema: a systematic review and meta-analysis. Ophthalmology 2015;122:1820-7. Crossref

47. Park YH, Shin JA, Han JH, Park YM, Yim HW. The association between chronic kidney disease and diabetic retinopathy: the Korea national health and nutrition examination survey 2008-2010. PLoS One 2015;10:e0125338. Crossref

48. Zhang H, Wang J, Ying GS, Shen L, Zhang Z. Diabetic retinopathy and renal function in Chinese type 2 diabetic patients. Int Urol Nephrol 2014;46:1375-81. Crossref

49. Penno G, Solini A, Zoppini G, et al. Rate and determinants of association between advanced retinopathy and chronic kidney disease in patients with type 2 diabetes: the Renal Insufficiency And Cardiovascular Events (RIACE) Italian multicenter study. Diabetes Care 2012;35:2317-23. Crossref

50. Rodríguez-Poncelas A, Mundet-Tudurí X, Miravet-Jiménez S, et al. Chronic kidney disease and diabetic retinopathy in patients with type 2 diabetes. PLoS One 2016;11:e0149448. Crossref

51. Gaedt Thorlund M, Borg Madsen M, Green A, Sjølie AK, Grauslund J. Is smoking a risk factor for proliferative diabetic retinopathy in type 1 diabetes? Ophthalmologica 2013;230:50-4. Crossref

52. Fu Y, Geng D, Liu H, Che H. Myopia and/or longer axial length are protective against diabetic retinopathy: a meta-analysis. Acta Ophthalmol 2016;94:346-52. Crossref

53. Wang X, Tang L, Gao L, Yang Y, Cao D, Li Y. Myopia and diabetic retinopathy: a systematic review and meta-analysis. Diabetes Res Clin Pract 2016;111:1-9. Crossref

54. Grant RW, Moore AF, Florez JC. Genetic architecture of type 2 diabetes: recent progress and clinical implications. Diabetes Care 2009;32:1107-14. Crossref

55. Grant SF, Thorleifsson G, Reynisdottir I, et al. Variant of transcription factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat Genet 2006;38:320-3. Crossref

56. Cauchi S, Froguel P. TCF7L2 genetic defect and type 2 diabetes. Curr Diab Rep 2008;8:149-55. Crossref

57. Ciccacci C, Di Fusco D, Cacciotti L, et al. TCF7L2 gene polymorphisms and type 2 diabetes: association with diabetic retinopathy and cardiovascular autonomic neuropathy. Acta Diabetol 2013;50:789-99. Crossref

58. Fu LL, Lin Y, Yang ZL, Yin YB. Association analysis of genetic polymorphisms of TCF7L2, CDKAL1, SLC30A8, HHEX genes and microvascular complications of type 2 diabetes mellitus [in Chinese]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 2012;29:194-9.

59. Ding Y, Hu Z, Yuan S, Xie P, Liu Q. Association between transcription factor 7-like 2 rs7903146 polymorphism and diabetic retinopathy in type 2 diabetes mellitus: A meta-analysis. Diab Vasc Dis Res 2015;12:436-44. Crossref

60. Choi HJ, Lee DH, Jeon HJ, Kim DS, Lee YH, Oh T. Transcription factor 7-like 2 (TCF7L2) gene polymorphism rs7903146 is associated with stroke in type 2 diabetes patients with long disease duration. Diabetes Res Clin Pract 2014;103:e3-6. Crossref

61. Ma J, Li Y, Zhou F, Xu X, Guo G, Qu Y. Meta-analysis of association between the Pro12Ala polymorphism of the peroxisome proliferator–activated receptor-gamma2 gene and diabetic retinopathy in Caucasians and Asians. Mol Vis 2012;18:2352-60.

62. Peng D, Wang J, Zhang R, et al. C-reactive protein genetic variant is associated with diabetic retinopathy in Chinese patients with type 2 diabetes. BMC Endocr Disord 2015;15:8. Crossref

63. Benjamin EJ, Dupuis J, Larson MG, et al. Genome-wide association with select biomarker traits in the Framingham Heart Study. BMC Med Genet 2007;8 Suppl 1:11S. Crossref

64. Klein BE, Knudtson MD, Tsai MY, Klein R. The relation of markers of inflammation and endothelial dysfunction to the prevalence and progression of diabetic retinopathy: Wisconsin epidemiologic study of diabetic retinopathy. Arch Ophthalmol 2009;127:1175-82. Crossref

65. Lim LS, Tai ES, Mitchell P, et al. C-reactive protein, body mass index, and diabetic retinopathy. Invest Ophthalmol Vis Sci 2010;51:4458-63. Crossref

66. Muni RH, Kohly RP, Lee EQ, Manson JE, Semba RD, Schaumberg DA. Prospective study of inflammatory biomarkers and risk of diabetic retinopathy in the diabetes control and complications trial. JAMA Ophthalmol 2013;131:514-21. Crossref

67. Nguyen TT, Alibrahim E, Islam FM, et al. Inflammatory, hemostatic, and other novel biomarkers for diabetic retinopathy: the multi-ethnic study of atherosclerosis. Diabetes Care 2009;32:1704-9. Crossref

68. van Hecke MV, Dekker JM, Nijpels G, et al. Inflammation and endothelial dysfunction are associated with retinopathy: the Hoorn Study. Diabetologia 2005;48:1300-6. Crossref

69. Peng D, Wang J, Zhang R, et al. Common variants in or near ZNRF1, COLEC12, SCYL1BP1 and API5 are associated with diabetic retinopathy in Chinese patients with type 2 diabetes. Diabetologia 2015;58:1231-8. Crossref

70. Fu YP, Hallman DM, Gonzalez VH, et al. Identification of diabetic retinopathy genes through a genome-wide association study among Mexican-Americans from Starr County, Texas. J Ophthalmol 2010;2010.pii: 861291.

71. Grassi MA, Tikhomirov A, Ramalingam S, Below JE, Cox NJ, Nicolae DL. Genome-wide meta-analysis for severe diabetic retinopathy. Hum Mol Genet 2011;20:2472-81. Crossref

The concepts underlying the pathogenesis of septic acute kidney injury (AKI) are complex. It is characterised by renal macro- and micro-circulatory disturbance, surge of inflammatory markers, and de-regulation of oxidative stress, followed by a bioenergetic adaptive response and controlled cell cycle arrest aimed at preventing cell death.1 Continuous renal replacement therapy is commonly performed in the critical care setting for patients with septic AKI. The use of low- or normal-volume continuous venovenous haemodialysis or haemofiltration, however, has failed to demonstrate any improvement of patient outcome in severe sepsis.2 3 Extracorporeal blood purification therapies have been proposed to improve the outcome for patients with severe sepsis with and without AKI. The underlying principle is the removal of excessive inflammatory mediators and/or bacterial toxins from the blood compartment in order to modulate the inflammatory response. This involves various techniques including haemoperfusion/haemoadsorption, high-adsorption haemofiltration, high-volume haemofiltration (HVHF), high cut-off (HCO) membrane haemofiltration/haemodialysis, plasma exchange, and coupled plasma filtration adsorption (CPFA) [Table 1]. These techniques are gaining popularity in Europe and Japan. This overview discusses the concept and latest advances in blood purification for sepsis and septic AKI.

During sepsis, triacylated peptides, diacylated peptides, or lipopolysaccharides (LPS) are released by pathogens, and are recognised by the Toll-like receptors located on the surface of antigen-presenting cells.4 5 Toll-like receptors also recognise locally produced damage-associated molecular patterns (DAMPs) from ischaemic renal tissue and circulating DAMPs released from extensive extrarenal tissue damage in sepsis.6 This triggers the activation of leukocytes, endothelial cells, and epithelial cells that release more inflammatory mediators such as tumour necrosis factor–alpha (TNF-α), interleukin-1 (IL-1), IL-6, IL-8 and IL-10, causing cellular and tissue damage.7 8 This is called a ‘cytokine storm’, and can also occur in non-infectious conditions such as severe trauma, extensive burns, acute necrotising pancreatitis, and post–cardiac arrest. A cytokine storm per se, in the absence of life-threatening triggering factors, can induce haemodynamic instability and multi-organ failure as illustrated by Suntharalingam et al.9 Moreover, immunoparalysis might occur after a cytokine storm and contribute to severe secondary nosocomial infections.10 As demonstrated in a postmortem by Boomer et al,11 patients who die of severe sepsis have biochemical and immunohistochemical findings consistent with immunosuppression. This gives rise to the concept of immunomodulation in sepsis. Low-dose steroid administration has been shown to improve septic shock reversal but is not associated with any survival benefits and is currently out of favour.12 13 The clinical benefit of intravenous immunoglobulins and anti–TNF-α in the treatment of severe sepsis is controversial and inconclusive.14 15 Blood purification may offer non-specific clearance of inflammatory mediators and/or microbial toxins and thus help to restore immune homeostasis. Five theories have been proposed to explain the potential benefit of blood purification in sepsis. First, Ronco et al16 proposed the “cytokine peak concentration hypothesis” and suggested that eliminating the peaks in cytokine blood concentration during the early phase of sepsis could stop the inflammatory cascade, limit organ damage, and consequently decrease the incidence of multi-organ failure syndrome. Second, Honoré and Matson17 proposed the “threshold immunomodulation hypothesis” that indicated cytokines will equilibrate between the blood and tissue compartments. This provided an explanation for the clinical benefit of blood purification techniques even without any significant changes in cytokine level within the blood compartment. Third, Di Carlo and Alexander18 proposed the “mediator delivery hypothesis” and suggested that high-volume fluid replacement during haemofiltration might promote lymphatic flow and displace inflammatory mediators to the blood compartment, making them available for removal. Fourth, Peng et al19 suggested that blood purification therapies could act directly at the cellular level to restore immune function. Finally, Rimmelé and Kellum20 proposed the “cytokinetic model” which indicated that blood purification techniques remove cytokines from the blood compartment and widen the cytokine/chemokine concentration gradient between blood and infected tissue. This improves leukocyte trafficking towards the infective foci, and thus promotes bacterial killing.

This technique binds toxins and other mediators in the extracorporeal circuit and removes them from the blood compartment.20 The sorbents, which consist of microfibres or resin-covered beads, are normally contained in cartridges that are placed in series within the extracorporeal circuit. They have a selective or non-selective binding capacity for cytokines, chemokines, super-antigens, or endotoxins by means of hydrophobic interaction, van der Waals forces, or ionic interactions.20 Initial clinical applications were complicated by severe thrombocytopenia and leukopenia but these were subsequently managed using a biocompatible coating.

Polymyxin B (PMX)–immobilised fibre column haemoperfusion (Toraymyxin, Toray Industries, Tokyo, Japan) is the most commonly used approach, and has been used for the treatment of septic shock since 1994 in Japan and since 2002 in Europe. It has gained popularity worldwide in recent years, especially after the landmark EUPHAS (Early Use of Polymyxin B Hemoperfusion in Abdominal Sepsis) study.21 The PMX is a group of cyclic cationic polypeptide antibiotics derived from Bacillus polymyxa. Endotoxins are heat and pH stable, and thus can be difficult to remove from protein-rich solutions such as blood. The PMX is capable of binding and neutralising endotoxins. Nephrotoxicity and neurotoxicity, however, are very common and thus limit their clinical use.22 To overcome this problem, PMX is immobilised onto polystyrene fibres that effectively remove endotoxin without leaching. The blood is perfused at a rate of 80 to 100 mL/min through a PMX-immobilised fibre column. Anticoagulation is achieved using unfractionated heparin, low-molecular-weight heparin, or the protease inhibitor nafamostat mesylate. Treatment usually lasts for 2 to 27 hours once or in some patients up to 4 times, depending on the clinical response. Three meta-analyses (approximately 1000 patients) were published before 2015: Studies by Mitaka and Tomita23 (17 studies, 975 patients) and Cruz et al24 (28 studies, 1425 patients) included both randomised controlled trials (RCTs) and observational studies. When reported, Gram-negative infections were identified in approximately 70% of patients (range, 37.9%-100% in individual studies). In general, PMX treatment led to significant haemodynamic improvement with a reduction in the use of inotropic agents/vasopressors in patients with sepsis. Moreover, it was associated with a decreased endotoxin level, modulation of inflammatory markers, and improvement of the PaO₂/FiO₂ ratio (ratio of the partial pressure of oxygen in arterial blood to the inspired oxygen fraction) in most included studies.23 24 Treatment by PMX significantly reduced 28-day mortality compared with conventional therapy. The meta-analysis by Zhou et al25 (8 studies, 370 patients) included RCTs only and focused on mortality, and showed significant survival benefit compared with conventional treatment. Only a few clinically important adverse effects were reported during PMX haemoperfusion, including cartridge clotting, hypotension, and hypersensitivity. Nonetheless, the largest multicentre RCT (232 patients) testing the performance of PMX haemoperfusion in peritonitis-induced septic shock was published in April 2015, and reported contrasting findings.26 No significant differences in 28-day mortality (27.7% in PMX-treated group vs 19.5% in controls; P=0.14), haemodynamic patterns, or organ failure evolution were observed. This negative result was similar to a large retrospective study (642 patients) by Iwagami et al27 who examined the effect of postoperative PMX haemoperfusion on peritonitis-induced septic shock. Patients treated with one or two PMX haemoperfusion sessions showed similar mortality at day 28 (17%) to propensity-matched patients without PMX treatment (16.3%). EUPHRATES (safety and efficacy of PMX haemoperfusion for septic shock study), a very large multicentre US-based phase III trial in patients with confirmed endotoxaemia, is currently underway and results should be available after July 2017.28 Based on current evidence, the clinical benefit of PMX haemoperfusion in Gram-negative sepsis is unclear. Moreover, the cost of individual haemoperfusion cartridges is very high (approximately HK$40 000 per cartridge) and limits its clinical use in local settings. Currently, PMX-immobilised fibre column haemoperfusion is not available in Hong Kong.

The MATISSE-Fresenius system (Fresenius SE, Bad Homburg, Germany) binds endotoxins to human albumin. The extracorporeal circuit is maintained by the Fresenius haemoadsorption machine using the MATISSE haemoadsorber that contains human serum albumin immobilised on polymethacrylate beads. Trends in the improvement of morbidity and organ dysfunction were reported in initial non-randomised studies,29 30 although a subsequent multicentre RCT could not identify any significant clinical benefit, which then limited its clinical use.31 Currently, the MATISSE-Fresenius system is not available in Hong Kong.

The Alteco LPS Adsorber (Alteco Medical AB, Lund, Sweden) captures endotoxins using specially designed synthetic peptides. This device was launched in 2006. Tailor-made synthetic peptides with a high affinity for endotoxins are attached to the surface of the polyethylene plates using a covalent bonding technique. Clinical experience with this device is scarce, and is limited mainly to case reports and case series.32 33 34 In general, these case series report a shorter vasopressor infusion duration in adsorber-treated patients compared with controls. Only one underpowered RCT has been published by local investigators.35 The study was terminated early and showed no significant clinical benefit (disease severity score, vasopressor use, length of study, and 28-day mortality) following the addition of the Alteco LPS Adsorber to conventional therapy in patients who had intra-abdominal sepsis with shock.35 The side-effect profile of this novel device was acceptable but a recent ex-vivo experimental study showed that the Alteco LPS Adsorber could not achieve acceptable LPS clearance in serum, heparinised plasma, or whole blood.36 Therefore, the potential benefit of the Alteco LPS Adsorber in sepsis is not clear.

Several cytokine-absorbing columns have been tested in animal studies, showing excellent adsorption rates for inflammatory cytokines such as TNF-α, IL-1β, IL-6, and IL-8.37 Human data are limited to case reports and case series.38 39 40 CytoSorb (CytoSorbents Corporation; Monmouth Junction [NJ], US) is a novel synthetic haemabsorption column that targets inflammatory mediators.41 It is currently the only European-approved extracorporeal device for cytokine haemoadsorption. Case reports show good cytokine clearance and haemodynamic improvement with this device.41 42 43 44 Further studies focused on clinically relevant endpoints are highly recommended. CytoSorb is not available in Hong Kong.

The AN69 and polymethylmethacrylate (PMMA) membrane haemofilters are the currently available options for performing high-adsorption haemofiltration in septic patients. Both have a high cytokine adsorption capacity but surface treatment can further modify their haemoadsorptive properties.45 46 47

oXiris (Gambro Hospal, Stockholm, Sweden) is an AN69-based membrane haemofilter that is surface-treated with a polyethyleneimine and grafted with heparin (Table 2). The AN69 core membrane has superior cytokine-binding capacity compared with the traditional polysulphone membrane. Surface treatment with polyethyleneimine enhances endotoxin capture,48 while heparin coating reduces membrane thrombogenicity, and prolongs the filter life and improves efficiency. A case-control study by Shum et al49 involving Gram-negative septicaemic patients (n=6) showed that oXiris continuous venovenous haemofiltration (CVVH) was associated with a greater reduction in Sequential Organ Failure Assessment score compared with conventional polysulphone-based CVVH (n=24). Subsequent large case series (n=40) suggested that oXiris treatment had a positive effect on haemodynamics with a reduction in cytokine levels (IL-6).50 Treatment usually lasts for 72 hours (manufacturer’s recommendation) and costs approximately HK$8000 per haemofilter. As of mid 2015, at least five public hospitals in Hong Kong have clinical experience with oXiris haemofilters in the treatment of septic shock. A large-scale RCT will be necessary to determine the potential benefit of this device, however.

The PMMA membrane has a higher cytokine adsorption capacity than the traditional polyacrylonitrile and polysulphone membrane.51 Membrane binding site saturation is one of the main concerns during treatment involving highly adsorptive haemofiltration. The PMMA haemofilter can maintain its cytokine adsorption capacity for at least 24 hours after being changed.52 The initial clinical experience of PMMA continuous haemodiafiltration in the treatment of sepsis is encouraging, with significant haemodynamic improvement and potential survival benefit.53 54 A local RCT (Australian New Zealand Clinical Trial Registry ACTRN12611000652976) that is aimed at investigating the clinical benefit of PMMA-based CVVH in patients with septic shock and AKI is currently underway. Treatment usually lasts for 24 to 48 hours and costs approximately HK$300 per haemofilter.

In 2002, HVHF was defined as >35 mL/kg/h, based on recommendations from the Acute Dialysis Quality Initiative Workgroup.55 Nonetheless in clinical practice, 35 mL/kg/h is not that high and can be achieved with ease, especially in those with low body weight. To clarify this issue, Honore et al56 defined continuous HVHF as 50 to 70 mL/kg/h, and 100 to 120 mL/kg/h for 4 to 8 hours followed by conventional CVVH as pulse HVHF. In addition, HVHF is regarded as effective blood purification therapy because circulating inflammatory mediators are mostly water-soluble and range between 5 kDa and 60 kDa. They are more effectively removed by convective means than by diffusion techniques. Moreover, haemofilter membranes have some adsorptive properties that allow the removal of mediators with a molecular weight higher than the membrane cut-off point. It is clear that conventional haemofiltration with low ultrafiltration rates is ineffective for cytokine removal.2 3 Increasing the ultrafiltration flow rate can increase the adsorption capacity of the haemofilter because of its effect on transmembrane pressure (greater membrane site recruitment) and the exposure of more available adsorptive surface area.57 Only two RCTs that investigated the potential benefit of HVHF over conventional CVVH in septic patients were available before the publication of the landmark trial (high-volume versus standard-volume haemofiltration for septic shock patients with acute kidney injury study) in 2013.57 58 Cole et al57 performed the first randomised crossover clinical trial that involved 11 patients with septic shock and multi-organ failure. Patients were assigned to either 8 hours of HVHF (6 L/h) or 8 hours of standard CVVH (1 L/h) in a random order. The results showed that HVHF was associated with a greater reduction in vasopressor use. A study by Boussekey et al58 (HVHF 65 mL/kg/h vs control 35 mL/kg/h; n=20) yielded similar findings and showed no survival benefit of HVHF over conventional CVVH. Multiple non-randomised studies showed decreased mortality with HVHF for septic shock patients but most of the studies were relatively small.59 60 61 Despite the initially encouraging results, HVHF has not gained in popularity because the use of a large volume of ultrapure replacement solution equates to significant increases in treatment cost, risk of severe electrolyte disturbance, and nursing workload. The landmark IVORIE study was published in 2013.62 This multicentre RCT involved 140 critically ill septic shock patients who were randomised to receive either HVHF at 70 mL/kg/h or standard CVVH treatment at 35 mL/kg/h. It showed neither significant survival benefit nor haemodynamic improvement for HVHF compared with standard treatment. Subsequently two meta-analyses (4 studies with approximately 500 patients) published in 2014 concluded that neither HVHF nor pulse HVHF offered any added clinical benefit when compared with standard-volume haemofiltration.63 64 Therefore, the routine use of HVHF for treatment of sepsis is not recommended.

Inflammatory mediators are relatively large (TNF-α: 17 kDa, IL-6: 26 kDa, and IL-8: 8 kDa), and are classified as middle molecules. The conventional high-flux haemofilter has a cut-off point at approximately 20 kDa and is unlikely to achieve good cytokine clearance.2 3 The nominal cut-off point for HCO membranes ranges from 60 to 150 kDa and the clinical cut-off point in blood ranges from 40 to 100 kDa.65 This can greatly increase the sieving coefficients of various inflammatory mediators at the expense of loss of albumin (66 kDa), antithrombin-III (60 kDa), protein C (62 kDa), and many other vital proteins. Reducing the pore size slightly can limit vital protein loss but also decrease cytokine removal. Ex-vivo studies showed that HCO haemofiltration displayed the greatest consistency in cytokine removal when compared with standard haemofiltration.66 The CPFA and haemoadsorption appeared to offer a similar level of cytokine clearance to the HCO technique. Albumin loss was comparable between HCO haemofiltration, HCO haemodialysis, and HCO haemodiafiltration.66 Morgera et al67 published the first study on the use of HCO haemofiltration among septic shock patients and showed good IL-6 (but not TNF-α) clearance. Subsequently, Morgera et al68 conducted an RCT that involved 30 septic AKI patients who were randomised to HCO or conventional haemofiltration. The HCO group showed a significant decline in vasopressor use and cytokine level.68 The largest RCT was the High Cut-Off Continuous Veno-venous Hemodialysis (CVVHD) in Patients Treated for Acute Renal Failure After Systemic Inflammatory Response Syndrome (SIRS)/Septic Shock (HICOSS) study.69 The estimated sample size was 120 patients but the study was terminated early because of a lack of difference between the groups after 81 patients had been recruited. There was no difference in 28-day mortality, vasopressor use, duration of mechanical ventilation, length of stay in intensive care unit, or albumin level between the groups.69 This underpowered RCT (due to premature termination) cannot provide a clear answer about the potential benefit of HCO haemofiltration/haemodialysis in septic patients and a further large-scale prospective RCT is recommended. Only the septeX (Gambro Hospal, Stockholm, Sweden) and EMiC 2 (Fresenius SE, Bad Homburg, Germany) HCO haemofilters are available in Hong Kong (Table 2). Treatment usually lasts for 24 to 72 hours and costs approximately HK$8000 per HCO haemofilter.

The nominal cut-off point for the plasma filter ranges from 400 to 800 kDa and therefore can achieve good cytokine removal with significant albumin loss (Table 2). Only three RCTs have been published to date. Busund et al70 published the largest RCT involving 106 adult septic patients randomised to receive either two sessions of plasmapheresis or standard therapy. Plasmapheresis offered better 28-day survival compared with the control group (67% vs 47%). Studies by Reeves et al71 and Long et al72 showed no survival benefit, however. Therefore, the debate regarding the benefit of plasmapheresis in sepsis continues. One important drawback of plasmapheresis is the significant loss of albumin, fibrinogen, antithrombin, and immunoglobulin that takes a long time to regenerate in the absence of post-treatment replacement.73 This problem can be resolved with the use of CPFA (Lynda, Bellco, Mirandola, Italy). This CPFA therapy comprises a plasma filter, a non-selective hydrophobic resin cartridge with high affinity for inflammatory mediators, and a high-flux haemofilter for convective solute removal (Fig).74 Only filtrated plasma has direct contact with the sorbents that have no biocompatibility problems when compared with direct haemoperfusion. Treatment lasts for approximately 10 hours and requires cartridge changes due to saturation problems. Livigni et al75 published the only multicentre RCT focused on patients with septic shock. Patients were randomised to standard treatment with or without CPFA. The CPFA therapy was performed daily for 5 days and lasted at least 10 hours/day. The estimated sample size was 330 patients but the study was terminated early on the grounds of futility after 192 patients had been recruited. No significant benefits for mortality, organ dysfunction, or intensive care unit stay were observed. Therefore, based on the available evidence, the routine use of CPFA for treatment of septic shock is not recommended. The CPFA is currently not available in Hong Kong.

Building on the concept of excessive inflammatory mediator release, blood purification techniques have emerged as an adjunctive therapy for patients with severe sepsis and septic AKI. They are effective in clearing endotoxin or inflammatory mediators and are well tolerated. Despite initially promising results, most blood purification techniques have not provided any sustainable mortality benefits. In severe sepsis, source control, early appropriate antibiotics, and haemodynamic support are the three most important treatment components.76 As a supportive treatment, blood purification techniques may not significantly affect patient mortality. Since the outcome for septic patients has improved over time, much larger sample sizes will be needed to detect the relatively small effects of these new therapies on sepsis.77 Large-scale, well-designed, prospective RCTs are the way forward. The application of these novel techniques should be individualised but more specific recommendations must await further evidence.

All authors have disclosed no conflicts of interest.

1. Shum HP, Yan WW, Chan TM. Recent knowledge on the pathophysiology of septic acute kidney injury: A narrative review. J Crit Care 2016;31:82-9. Crossref

2. Cole L, Bellomo R, Hart G, et al. A phase II randomized, controlled trial of continuous hemofiltration in sepsis. Crit Care Med 2002;30:100-6. Crossref

3. Payen D, Mateo J, Cavaillon JM, et al. Impact of continuous venovenous hemofiltration on organ failure during the early phase of severe sepsis: a randomized controlled trial. Crit Care Med 2009;37:803-10. Crossref

4. Prince LR, Whyte MK, Sabroe I, Parker LC. The role of TLRs in neutrophil activation. Curr Opin Pharmacol 2011;11:397-403. Crossref

5. Nakayama H, Kurokawa K, Lee BL. Lipoproteins in bacteria: structures and biosynthetic pathways. FEBS J 2012;279:4247-68. Crossref

6. Zarbock A, Gomez H, Kellum JA. Sepsis-induced acute kidney injury revisited: pathophysiology, prevention and future therapies. Curr Opin Crit Care 2014;20:588-95. Crossref

7. Angus DC, van der Poll T. Severe sepsis and septic shock. N Engl J Med 2013;369:2063. Crossref

8. Gomez H, Ince C, De Backer D, et al. A unified theory of sepsis-induced acute kidney injury: inflammation, microcirculatory dysfunction, bioenergetics, and the tubular cell adaptation to injury. Shock 2014;41:3-11. Crossref

9. Suntharalingam G, Perry MR, Ward S, et al. Cytokine storm in a phase 1 trial of the anti-CD28 monoclonal antibody TGN1412. N Engl J Med 2006;355:1018-28. Crossref

10. Limaye AP, Kirby KA, Rubenfeld GD, et al. Cytomegalovirus reactivation in critically ill immunocompetent patients. JAMA 2008;300:413-22. Crossref

11. Boomer JS, To K, Chang KC, et al. Immunosuppression in patients who die of sepsis and multiple organ failure. JAMA 2011;306:2594-605. Crossref

12. Wang C, Sun J, Zheng J, et al. Low-dose hydrocortisone therapy attenuates septic shock in adult patients but does not reduce 28-day mortality: a meta-analysis of randomized controlled trials. Anesth Analg 2014;118:346-57. Crossref

13. Kalil AC, Sun J. Low-dose steroids for septic shock and severe sepsis: the use of Bayesian statistics to resolve clinical trial controversies. Intensive Care Med 2011;37:420-9. Crossref

14. Alejandria MM, Lansang MA, Dans LF, Mantaring JB 3rd. Intravenous immunoglobulin for treating sepsis, severe sepsis and septic shock. Cochrane Database Syst Rev 2013;(9):CD001090.

15. Bernard GR, Francois B, Mira JP, et al. Evaluating the efficacy and safety of two doses of the polyclonal anti-tumor necrosis factor-alpha fragment antibody AZD9773 in adult patients with severe sepsis and/or septic shock: randomized, double-blind, placebo-controlled phase IIb study. Crit Care Med 2014;42:504-11. Crossref

16. Ronco C, Tetta C, Mariano F, et al. Interpreting the mechanisms of continuous renal replacement therapy in sepsis: the peak concentration hypothesis. Artif Organs 2003;27:792-801. Crossref

17. Honoré PM, Matson JR. Extracorporeal removal for sepsis: Acting at the tissue level—the beginning of a new era for this treatment modality in septic shock. Crit Care Med 2004;32:896-7. Crossref

18. Di Carlo JV, Alexander SR. Hemofiltration for cytokine-driven illnesses: the mediator delivery hypothesis. Int J Artif Organs 2005;28:777-86.

19. Peng Z, Singbartl K, Simon P, et al. Blood purification in sepsis: a new paradigm. Contrib Nephrol 2010;165:322-8. Crossref

20. Rimmelé T, Kellum JA. Clinical review: blood purification for sepsis. Crit Care 2011;15:205. Crossref

21. Cruz DN, Antonelli M, Fumagalli R, et al. Early use of polymyxin B hemoperfusion in abdominal septic shock: the EUPHAS randomized controlled trial. JAMA 2009;301:2445-52. Crossref

22. Shoji H. Extracorporeal endotoxin removal for the treatment of sepsis: endotoxin adsorption cartridge (Toraymyxin). Ther Apher Dial 2003;7:108-14. Crossref

23. Mitaka C, Tomita M. Polymyxin B–immobilized fiber column hemoperfusion therapy for septic shock. Shock 2011;36:332-8. Crossref

24. Cruz DN, Perazella MA, Bellomo R, et al. Effectiveness of polymyxin B–immobilized fiber column in sepsis: a systematic review. Crit Care 2007;11:R47. Crossref

25. Zhou F, Peng Z, Murugan R, Kellum JA. Blood purification and mortality in sepsis: a meta-analysis of randomized trials. Crit Care Med 2013;41:2209-20. Crossref

26. Payen DM, Guilhot J, Launey Y, et al. Early use of polymyxin B hemoperfusion in patients with septic shock due to peritonitis: a multicenter randomized control trial. Intensive Care Med 2015;41:975-84. Crossref

27. Iwagami M, Yasunaga H, Doi K, et al. Postoperative polymyxin B hemoperfusion and mortality in patients with abdominal septic shock: a propensity-matched analysis. Crit Care Med 2014;42:1187-93. Crossref

28. Klein DJ, Foster D, Schorr CA, Kazempour K, Walker PM, Dellinger RP. The EUPHRATES trial (Evaluating the Use of Polymyxin B Hemoperfusion in a Randomized controlled trial of Adults Treated for Endotoxemia and Septic shock): study protocol for a randomized controlled trial. Trials 2014;15:218. Crossref

29. Ullrich H, Jakob W, Frohlich D, et al. A new endotoxin adsorber: first clinical application. Ther Apher 2001;5:326-34. Crossref

30. Staubach KH, Boehme M, Zimmermann M, Otto V. A new endotoxin adsorption device in Gram-negative sepsis: use of immobilized albumin with the MATISSE adsorber. Transfus Apher Sci 2003;29:93-8. Crossref

31. Reinhart K, Meier-Hellmann A, Beale R, et al. Open randomized phase II trial of an extracorporeal endotoxin adsorber in suspected Gram-negative sepsis. Crit Care Med 2004;32:1662-8. Crossref

32. Kulabukhov VV. Use of an endotoxin adsorber in the treatment of severe abdominal sepsis. Acta Anaesthesiol Scand 2008;52:1024-5. Crossref

33. Yaroustovsky M, Abramyan M, Popok Z, et al. Preliminary report regarding the use of selective sorbents in complex cardiac surgery patients with extensive sepsis and prolonged intensive care stay. Blood Purif 2009;28:227-33. Crossref

34. Ala-Kokko TI, Laurila J, Koskenkari J. A new endotoxin adsorber in septic shock: observational case series. Blood Purif 2011;32:303-9. Crossref

35. Shum HP, Leung YW, Lam SM, Chan KC, Yan WW. Alteco endotoxin hemoadsorption in Gram-negative septic shock patients. Indian J Crit Care Med 2014;18:783-8. Crossref

36. Harm S, Falkenhagen D, Hartmann J. Endotoxin adsorbents in extracorporeal blood purification: do they fulfill expectations? Int J Artif Organs 2014;37:222-32. Crossref

37. Taniguchi T. Cytokine adsorbing columns. Contrib Nephrol 2010;166:134-41. Crossref

38. Kellum JA, Venkataraman R, Powner D, Elder M, Hergenroeder G, Carter M. Feasibility study of cytokine removal by hemoadsorption in brain-dead humans. Crit Care Med 2008;36:268-72. Crossref

39. Kobe Y, Oda S, Matsuda K, Nakamura M, Hirasawa H. Direct hemoperfusion with a cytokine-adsorbing device for the treatment of persistent or severe hypercytokinemia: a pilot study. Blood Purif 2007;25:446-53. Crossref

40. Tsuchida K, Takemoto Y, Sugimura K, Yoshimura R, Nakatani T. Direct hemoperfusion by using Lixelle column for the treatment of systemic inflammatory response syndrome. Int J Mol Med 2002;10:485-8. Crossref

41. Morris C, Gray L, Giovannelli M. Early report: The use of CytoSorb haemabsorption column as an adjunct in managing severe sepsis: initial experiences, review and recommendations. J Intensive Care Soc 2015;16:257-64. Crossref

42. Hinz B, Jauch O, Noky T, Friesecke S, Abel P, Kaiser R. CytoSorb, a novel therapeutic approach for patients with septic shock: a case report. Int J Artif Organs 2015;38:461-4. Crossref

43. Wiegele M, Krenn CG. CytoSorb in a patient with Legionella pneumonia–associated rhabdomyolysis: a case report. ASAIO J 2015;61:e14-6. Crossref

44. Hetz H, Berger R, Recknagel P, Steltzer H. Septic shock secondary to beta-hemolytic streptococcus-induced necrotizing fasciitis treated with a novel cytokine adsorption therapy. Int J Artif Organs 2014;37:422-6. Crossref

45. Bouman CS, van Olden RW, Stoutenbeek CP. Cytokine filtration and adsorption during pre- and postdilution hemofiltration in four different membranes. Blood Purif 1998;16:261-8. Crossref

46. De Vriese AS, Colardyn FA, Philippé JJ, Vanholder RC, De Sutter JH, Lameire NH. Cytokine removal during continuous hemofiltration in septic patients. J Am Soc Nephrol 1999;10:846-53.

47. Hirasawa H, Oda S, Matsuda K. Continuous hemodiafiltration with cytokine-adsorbing hemofilter in the treatment of severe sepsis and septic shock. Contrib Nephrol 2007;156:365-70. Crossref

48. Mitzner S, Schneidewind J, Falkenhagen D, Loth F, Klinkmann H. Extracorporeal endotoxin removal by immobilized polyethylenimine. Artif Organs 1993;17:775-81. Crossref

49. Shum HP, Chan KC, Kwan MC, Yan WW. Application of endotoxin and cytokine adsorption haemofilter in septic acute kidney injury due to Gram-negative bacterial infection. Hong Kong Med J 2013;19:491-7. Crossref

50. Turani F, Candidi F, Barchetta R, et al. Continuous renal replacement therapy with the adsorbent membrane oXiris in septic patients: a clinical experience. Crit Care 2013;17 Suppl 2:63. Crossref

51. Matsuda K, Hirasawa H, Oda S, Shiga H, Nakanishi K. Current topics on cytokine removal technologies. Ther Apher 2001;5:306-14. Crossref

52. Nakada TA, Hirasawa H, Oda S, Shiga H, Matsuda K. Blood purification for hypercytokinemia. Transfus Apher Sci 2006;35:253-64. Crossref

53. Nakada TA, Oda S, Matsuda K, et al. Continuous hemodiafiltration with PMMA hemofilter in the treatment of patients with septic shock. Mol Med 2008;14:257-63. Crossref

54. Matsuda K, Moriguchi T, Harii N, Yanagisawa M, Harada D, Sugawara H. Comparison of efficacy between continuous hemodiafiltration with a PMMA high-performance membrane dialyzer and a PAN membrane hemofilter in the treatment of septic shock patients with acute renal failure. Contrib Nephrol 2011;173:182-90. Crossref

55. Kellum JA, Mehta RL, Angus DC, Palevsky P, Ronco C, ADQI Workgroup. The first international consensus conference on continuous renal replacement therapy. Kidney Int 2002;62:1855-63. Crossref

56. Honore PM, Joannes-Boyau O, Boer W, Collin V. High-volume hemofiltration in sepsis and SIRS: current concepts and future prospects. Blood Purif 2009;28:1-11. Crossref

57. Cole L, Bellomo R, Journois D, Davenport P, Baldwin I, Tipping P. High-volume haemofiltration in human septic shock. Intensive Care Med 2001;27:978-86. Crossref

58. Boussekey N, Chiche A, Faure K, et al. A pilot randomized study comparing high and low volume hemofiltration on vasopressor use in septic shock. Intensive Care Med 2008;34:1646-53. Crossref

59. Cornejo R, Downey P, Castro R, et al. High-volume hemofiltration as salvage therapy in severe hyperdynamic septic shock. Intensive Care Med 2006;32:713-22. Crossref

60. Joannes-Boyau O, Rapaport S, Bazin R, Fleureau C, Janvier G. Impact of high volume hemofiltration on hemodynamic disturbance and outcome during septic shock. ASAIO J 2004;50:102-9. Crossref

61. Piccinni P, Dan M, Barbacini S, et al. Early isovolaemic haemofiltration in oliguric patients with septic shock. Intensive Care Med 2006;32:80-6. Crossref

62. Joannes-Boyau O, Honoré PM, Perez P, et al. High-volume versus standard-volume haemofiltration for septic shock patients with acute kidney injury (IVOIRE study): a multicentre randomized controlled trial. Intensive Care Med 2013;39:1535-46. Crossref

63. Lehner GF, Wiedermann CJ, Joannidis M. High-volume hemofiltration in critically ill patients: a systematic review and meta-analysis. Minerva Anestesiol 2014;80:595-609.

64. Clark E, Molnar AO, Joannes-Boyau O, Honoré PM, Sikora L, Bagshaw SM. High-volume hemofiltration for septic acute kidney injury: a systematic review and meta-analysis. Crit Care 2014;18:R7. Crossref

65. Haase M, Bellomo R, Morgera S, Baldwin I, Boyce N. High cut-off point membranes in septic acute renal failure: a systematic review. Int J Artif Organs 2007;30:1031-41.

66. Atan R, Crosbie D, Bellomo R. Techniques of extracorporeal cytokine removal: a systematic review of the literature. Blood Purif 2012;33:88-100. Crossref

67. Morgera S, Rocktäschel J, Haase M, et al. Intermittent high permeability hemofiltration in septic patients with acute renal failure. Intensive Care Med 2003;29:1989-95. Crossref

68. Morgera S, Haase M, Kuss T, et al. Pilot study on the effects of high cutoff hemofiltration on the need for norepinephrine in septic patients with acute renal failure. Crit Care Med 2006;34:2099-104. Crossref

69. Honore P, Beck W, editors. High cut-off continuous venovenous hemodialysis (CVVHD) in patients treated for acute renal failure after systemic inflammatory response syndrome (SIRS)/septic shock (HICOSS). Proceedings of the 10th Congress of the World Federation of Societies of Intensive and Critical Care Medicine (WFSICCM); 2009 Aug 28-Sep 1; Florence, Italy.

70. Busund R, Koukline V, Utrobin U, Nedashkovsky E. Plasmapheresis in severe sepsis and septic shock: a prospective, randomised, controlled trial. Intensive Care Med 2002;28:1434-9. Crossref

71. Reeves JH, Butt WW, Shann F, et al. Continuous plasmafiltration in sepsis syndrome. Plasmafiltration in Sepsis Study Group. Crit Care Med 1999;27:2096-104. Crossref

72. Long EJ, Shann F, Pearson G, Buckley D, Butt W. A randomised controlled trial of plasma filtration in severe paediatric sepsis. Crit Care Resusc 2013;15:198-204.

73. Stegmayr B, Abdel-Rahman EM, Balogun RA. Septic shock with multiorgan failure: from conventional apheresis to adsorption therapies. Semin Dial 2012;25:171-5. Crossref

74. Formica M, Inguaggiato P, Bainotti S, Wratten ML. Coupled plasma filtration adsorption. Contrib Nephrol 2007;156:405-10. Crossref

75. Livigni S, Bertolini G, Rossi C, et al. Efficacy of coupled plasma filtration adsorption (CPFA) in patients with septic shock: a multicenter randomised controlled clinical trial. BMJ Open 2014;4:e003536. Crossref

76. Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013;41:580-637. Crossref

77. Kaukonen KM, Bailey M, Suzuki S, Pilcher D, Bellomo R. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000-2012. JAMA 2014;311:1308-16. Crossref

Over the past 50 years, a number of dental public health measures and policies have been established by the government in Hong Kong to help improve the oral health of the population. Children have been the focus for many of these dental public health practices since the 1960s and these have included prevention strategies, oral health education, and the provision of the School Dental Care Service (SDCS).

Historical development of Hong Kong’s public health measures for children

Water fluoridation is one of the most successfully implemented dental public health measures in Hong Kong. The project was launched in 1961 and has remarkably reduced the prevalence of dental caries in Hong Kong.1 All areas with a centralised water supply are fluoridated. Prior to its implementation, the natural fluoride concentration of drinking water in Hong Kong was less than 0.13 parts per million (ppm). Several adjustments have been made to the water fluoride level in Hong Kong since its implementation: from 0.7 ppm for summer months and 0.9 ppm for winter months in 1961, to 1 ppm in 1967; then reduced to 0.7 ppm in 1978; and further to 0.5 ppm in 1988 because of concerns of an increased prevalence of dental fluorosis in the population.

In late 1979, the SDCS was introduced to provide prevention and dental treatment to primary schoolchildren in Hong Kong. The SDCS also aims to promote oral health by delivering oral health education to schoolchildren. Preschool children in Hong Kong are not routinely eligible for the SDCS. They receive oral health care and treatment largely from dentists working in the private sector. Oral health education for preschool children was introduced through the ‘Brighter Smiles for the New Generation’ programme by the government in 1993. This programme promotes oral health awareness by educating children aged under 6 years about good oral health–related behaviour. It also aims to increase their teachers’ and parents’ oral health care knowledge.

The community is served by registered professional oral health care personnel. The Faculty of Dentistry at the University of Hong Kong was established in 1981 and began training dentists and supporting dental personnel in the same year. More than one third of local practising dentists have been educated in Hong Kong.2 The number of practising dentists currently serving the community has increased to about 2310 personnel, a per capita ratio of 1:3125.3 Before 1973, there were only about 440 practising dentists, a per capita ratio of 1:9000. In the past, it was often only the economic affluent who sought treatment from private dental practitioners. For many others, teeth were considered a dispensable commodity.2 Although this situation has improved, access to dental care for the Hong Kong population remains inadequate.

Oral health care products (fluoridated toothpaste, mouth rinses, toothbrushes, and floss) are now widely available in Hong Kong.4 These easily accessible fluoride-containing products provide an additional benefit to the oral health of the Hong Kong population. Most locally available toothpastes have a fluoride concentration of 1000 to 1500 ppm, while those for children have 600 ppm. Mouth rinse with 0.05% sodium fluoride is also freely available, though its use among children is not common and it is not recommended for the very young.

Over the years, dentistry in Hong Kong has advanced and oral health care has greatly improved under several government public health policies. Although children have been the primary target of such initiatives, information and review of the effects of the changes and trends in the oral health of Hong Kong children are limited. This is in part due to the limited dissemination of findings, particularly those of earlier government reports. Understanding the trends and current oral health condition of Hong Kong children is important. It can provide a historical and epidemiological overview of dental activity and inform the planning of future public health measures, prevention, and services for children. It may also help set future oral health targets and specific goals.

This paper reviews all available oral health epidemiological data and information of Hong Kong children from published literature before 2014 through electronic database searches, supplemented with information accessed from government-archived oral health reports. Reference lists of articles retrieved from the electronic databases were hand-searched for any other articles that might provide information relevant to the objectives of this paper. Major oral health problems of Hong Kong children—including dental caries experience, periodontal health, enamel defects, malocclusion, and orthodontic treatment need—are described.

A number of population-based oral epidemiological studies involving children have been conducted in Hong Kong since the 1960s. Available epidemiological data regarding dental caries experience and the extent/severity of dental caries among Hong Kong children are summarised in Table 1.4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 These studies employed different sampling methods as criteria of assessment differed prior to the 1970s. More recent surveys have followed World Health Organization’s criteria for caries assessment.

To date, there are 13 population-based epidemiological studies reporting the prevalence and extent of dental caries experience among Hong Kong schoolchildren and adolescents. A remarkable decrease in caries experience and severity among schoolchildren and adolescents has been observed since the 1960s.

The earliest report of dental caries experience among schoolchildren and adolescents was carried out in 1960, a year before the implementation of water fluoridation.5 The epidemiological findings gave cause for concern. Almost all children aged 6 to 11 years who participated in the study (aged 6-8: 97.5%; aged 9-11: 93.4%) had dental caries. The mean number of decayed, missing, or filled permanent teeth (DMFT) in 6-8-year-olds and 9-11-year-olds was 2.7 and 4.4, respectively; the mean decayed, missing, or filled primary teeth (dmft) in 6-8- and 9-11-year-olds was 9.2 and 3.8, respectively. Decayed teeth constituted the major component (>90%) of the dmft/DMFT index in both dentitions, and extracted and filled teeth components were minimal. This signified that there were no systematic dental care services available at the time and with limited preventive measures.

In 1962, the second population-wide oral health survey was conducted using similar sampling methodology to the 1960 survey,6 1 year after the implementation of water fluoridation. There was a reported slight decrease in the prevalence of dental caries experience (for those aged 6-8 years: 96.4%; for those aged 9-11 years: 90.0%). The mean dmft and DMFT showed a significant decline of approximately 20% (mean DMFT: 2.2 for those aged 6-8 years and 3.5 for those aged 9-11 years; mean dmft: 8.4 for those aged 6-8 years and 3.1 for those aged 9-11 years). As in the 1960 survey, decayed teeth constituted the major component of the DMFT, showing limited change in the provision/usage of dental care services. As this survey was conducted a year after water fluoridation, the decrease in dental caries could not be fully explained by exposure to fluoridated water. No report on the difference in dmft/DMFT among children who had received fluoridated water for the whole time, intermittently, or not at all was provided.

The third oral health survey of Hong Kong was completed in 1968, 7 years after the introduction of water fluoridation.7 The sampling method differed to the earlier surveys. Subjects aged 3 to 54 years were selected. The dental caries experience in the primary dentition was 83.8% (mean dmft: 5.0) for 3-8-year-olds, whereas the dental caries experience in the permanent dentition was 67.2% (mean DMFT: 2.0) for 9-14-year-olds, and 84.0% (mean DMFT: 4.6) for 15-19-year-olds. It represented a significant decrease in both caries experience and its extent among children when compared with previous surveys. This favourable change was attributed to water fluoridation as there were no other widely available caries preventive measures or systematic dental care services in the 1960s.

In 1980, another population-based dental health survey collected baseline data for future evaluation of the SDCS.8 The majority of schoolchildren (aged 6-11 years) examined were caries-free in their permanent dentition. The mean DMFT was <1 for children aged <9 years and 1.5 for 11-year-olds. The caries experience in the primary dentition of the children was high, however. The mean dmft for 6-year-olds was 4.3. The number of extracted and filled teeth for both dentitions was low. More than 90% of decayed teeth were untreated, indicating the lack of utilisation of dental services and a very high unmet treatment need.

The epidemiological studies conducted in the 1980s after the commencement of SDCS to monitor the effect of fluoride on dental caries experience of Hong Kong schoolchildren and adolescents after over 20 years of water fluoridation showed a further decrease in caries experience and severity. The reported mean DMFT of schoolchildren and adolescents in these surveys ranged from 0.3 to 2.89 10 11 12 and the mean dmft was 2.2.11 The caries experience was >65% in primary dentition and 18.3% to 77% in permanent dentition.11 12 The major component of the DMFT of the children of this time was decayed teeth, demonstrating that there was still a high unmet need for dental services, although the SDCS was already enacted. Caries was mostly experienced in molar teeth for children aged 9 to 12 years.

The caries prevalence and extent among Hong Kong schoolchildren and adolescents continued to show improvements. Kwan13 reported the first survey of schoolchildren aged 13 and 15 years who joined the SDCS in 1992. The dental caries experience was approximately 55%, with a mean DMFT of 1.3 for 13-year-olds and 1.6 for 15-year-olds. The result corresponded to a study by Yiu et al14 that reported a caries experience of 51% and a mean DMFT of 1.4 among 15-year-olds. In 1992, Evans and Lo15 also studied the effects of the SDCS on the dental status of primary teeth among a sample of Chinese children aged 6 to 12 years. The caries experience was 68%, with dmft indices for 6-, 7-, 8-, 9-, and 10-year-olds being 2.8, 3.1, 2.9, 2.3, and 1.3, respectively. The ratio of decayed-to-filled teeth decreased from 3.2 at age 6 to 1.0 at age 9. The mean number of filled teeth was the major component of the dmft index in these surveys, indicating that many children had received dental care.

The Hong Kong SAR Government conducted a population-based survey of the oral health status of 12-year-old children in 2001 and 2011.16 17 More than one-third (37.8%) of the children in 2001 had a caries experience in their permanent dentition and in 2011, 22.6% of the children had a caries experience. The extent/severity of caries was low (mean DMFT: 0.8 in 2001, 0.4 in 2011). Most of the decay experience was attributed to the filled component. The proportion of untreated decay was also rather low, with only 5.4% reported to have untreated decayed teeth in 2011. This positive development in oral health was associated with reported better oral health knowledge and oral care habits in both parents and children. A large number of the participants claimed they had regular dental check-ups.

In the past 50 years, for Hong Kong schoolchildren and adolescents, the prevalence of dental caries experience in permanent teeth has reduced from more than 90% (in the 1960s) to approximately 50% in the 1980s/90s and to less than 25% currently. The mean number of DMFT has also declined from over 4 in the 1960s to approximately 2 in the 1980s/90s, and to less than 1 currently.

There were seven epidemiological studies reporting dental caries experience among preschool children. Improvement among preschool children is less since the 1960s, compared with improvements among schoolchildren and adolescents and it remains a considerable problem.

The earliest oral health survey that involved young children was the population-based survey conducted in 1968.7 The prevalence of dental caries experience among 3-8-year-old children at that time was over 80%, with a dmft of 5. One quarter of the primary teeth of the children were decayed. The second report of preschool children caries experience in Hong Kong was drawn up by Wei et al.18 Conducted between 1986 and 1988 among approximately 10% of 5-year-old children, the percentage of children with caries in their primary dentition was 63%. The mean dmft was 3.2. Dental caries experience was higher for children from socio-economically disadvantaged families. Over 70% of the children had never visited a dentist.

The caries experience of Hong Kong preschool children further decreased to about 50% in the late 1990s.4 The mean dmft of children (4-6-year-olds) was 1.6. More than 90% of the dmft score was attributed to decayed untreated teeth. Similar to the findings of Wei et al,18 the children’s caries experience was associated with underprivileged socio-economic background, and parental educational level, dental knowledge, and attitudes.

In the recent decade, no great changes among preschool children caries status have been observed. The caries prevalence in preschool children remains similar, with a reported prevalence of 35% to 51%.16 17 19 20 The extent/severity of caries, however, showed a slight increase when compared with the late 1990s (mean dmft of children aged 3-5 years ranged from 1.5 to 2.5). Over 90% of the decayed teeth of the children were untreated. Almost one tenth of the children presented with abscess, with a higher percentage reported in the 2011 survey than in the 2001 survey.16 17 These recent surveys also found that children’s caries experience was associated with their place of birth, socio-economic background, and dietary habits.

Detailed information about the periodontal status of Hong Kong children is less readily available when compared with information about dental caries experience. Different assessment criteria have been used to assess periodontal health among children and adolescents, making it difficult to compare surveys.

The earliest report of the periodontal health status of Hong Kong children (aged 3-19 years) was drawn up by Wong in 1968.7 Wong7 reported that “oral hygiene was only fair in 70% of the children” and that “over 60% of the children had inflamed gingiva, and material alba was found on over 90% of the teeth surfaces”. Inflamed gingiva (gingivitis) is the reversible and non-destructive form of periodontal disease. This suggested that the children had poor periodontal health, though the criteria of assessment were not defined.

Law8 provided more specific details about the periodontal health status of 5-14-year-old Hong Kong children. Approximately 85% were reported to have soft deposits (assume plaque), of whom approximately one (19.2%) in five had ‘intensive gingivitis’ (inflamed gingiva). Calculus was observed among over a quarter (26.4%) of the children, and the percentage of calculus deposits increased with age.

Epidemiological studies conducted in the late 1980s employed the Community Periodontal Index (CPI) to assess the periodontal health status. It is the standard epidemiological index for assessing periodontal health,21 and results of the surveys were comparable. The epidemiological studies reporting periodontal health status of Hong Kong children using the CPI are shown in Table 2.10 11 13 14 16 17 The majority reported periodontal health status of adolescents. Among adolescents (13-18 years old), two of three studies10 13 14 reported that less than 10% had ‘healthy’ periodontal status (CPI=0) and more than half had evidence of calculus in some parts of the mouth (CPI=2). This showed that the periodontal health of Hong Kong adolescents was unsatisfactory.

For Hong Kong schoolchildren, the first detailed report of their periodontal health status was conducted by Wong11 among 7882 primary schoolchildren (aged 6-11 years) in 1987. More than half of the sextants (3.9) of the children had healthy gingiva (CPI=0). Nonetheless, more than half (56.1%) of the children had bleeding gingiva (CPI=1) or calculus deposits (CPI=2).

Two population-based oral health surveys among schoolchildren (12-year-old children) were conducted by the government in 2001 and 2011.16 17 The 2011 survey showed an improvement in periodontal health status. More schoolchildren were found to have healthy gingiva (CPI=0) in 2011: 13.8% compared with 5.5% in the 2001 survey; and less children had calculus deposits (CPI=2) in 2011: 22.4% compared with 59.5% in the 2001 survey. Of note, more than half of their sextants had either bleeding gingiva or calculus deposits in 2001 and an average of two of the sextants had these problems in 2011.

Studies that applied the Visible Plaque Index to assess periodontal health of Hong Kong preschool children are shown in Table 3.16 17 22 Such Index was introduced by Ainamo and Bay23 as a standardised assessment of oral hygiene status. It is simple and reliable to use and has been employed in surveys as a proxy of gingival health, representing the site prevalence of ‘clearly visible dental plaque’ at the gingival margin. The oral hygiene of the preschool children in Hong Kong was poor.16 17 22 Almost all 5-year-old children (97%) had at least one site with ‘clearly visible dental plaque’.16 17 The mean percentage of tooth surfaces with visible dental plaque was 22.1% in 2011 and 23.5% in 2001. A study conducted among 531 children aged 3 to 4 years in 2009, however, reported that 49.7% of the tooth surfaces had visible plaque.22 In general, the gingival condition and tooth cleanliness of both schoolchildren and preschool children were unsatisfactory and required much improvement.

Dental fluorosis consequent to exposure to fluoride may develop during the formation of teeth in young children. Table 4 outlines the studies reporting dental fluorosis using Dean’s Index in Hong Kong to date.5 6 10 11 Epidemiological studies reporting dental fluorosis showed that the level was low in the 1960s.5 6 Nearly all examined children aged 6 to 11 years had ‘normal’ enamel. Less than 1% were assessed as having a ‘mild’ or ‘questionable’ degree of fluorosis in 1960 or having a ‘moderate’ or ‘questionable’ degree of fluorosis in 1962.

Dental fluorosis was reported to be more prevalent in the 1980s, particularly among adolescents.10 11 Over 80% of the 15-19-year-olds in the study by Lind et al10 exhibited signs of dental fluorosis; and approximately 50% of the 6-11-year-olds in the study by Wong11 showed various degrees of dental fluorosis.

King24 reported the prevalence of enamel defects among a random sample of 12-year-old children in Hong Kong. The prevalence of teeth with opacities was 99.6%; 82.8% had evidence of hypoplasia and 16.6% had discoloured teeth. The author believed that many of the enamel defects were likely to be related to dental fluorosis. The findings in the 1980s suggested a marked increase in the level of dental fluorosis since the introduction of water fluoridation in the 1960s and it was then advocated to lower the water fluoridated level.1 The prevalence declined considerably over the decades when the level of fluoride in the water supply was adjusted and lowered.25 At present the level of water fluoridation is optimal at 0.5 ppm. The trend of decreasing dental fluorosis prevalence was evident in the study by Evans and Stamm.26 The prevalence decreased from 64% to 47% across the cohort of children from older (aged 12) to younger (aged 7) born before and after reduction of fluoride level to 0.7 ppm in 1978.

The prevalence and severity of developmental defects of enamel (DDE) have also been studied. Cross-sectional surveys showed that the prevalence of diffuse opacities among random samples of 12-year-old children (based on maxillary incisors and assessed using standardised intra-oral photographs) declined from 89.3% in 1983 to 32.4% in 2001, but increased to 42.1% in 2010.27 The mouth prevalence of DDE among maxillary incisor teeth of the children also decreased from 92.1% in 1983 to 35.2% in 2001.28 Wong et al29 also reported the prevalence of DDE among Hong Kong 12-year-old children at 90% in a 2010 cohort study (89.5% had diffuse opacities, 8.6% demarcated opacities, and 1.8% hypoplasia), using the modified version of the DDE index by FDI (Fédération Dentaire Internationale) to diagnose DDE.30 The prevalence of molar incisor hypomineralisation was reported as 2.8% among Primary 6 Chinese schoolchildren in a 2006 retrospective study.31

Epidemiological data on malocclusion and orthodontic treatment need among Hong Kong children are scant. Relatively few surveys have been conducted since the 1960s but details of the available studies are shown in Table 5.7 8 10 11 32 33 The studies were heterogeneous in terms of criteria used to assess dentofacial anomalies that require orthodontic intervention and age of the sampled children. Comparison and description of estimates of orthodontic treatment are difficult.

Earlier epidemiological studies of malocclusion and orthodontic treatment need among Hong Kong children suggested that dentofacial deformities requiring treatment intervention was ≤20%. More recent studies suggest orthodontic treatment need to be closer to 40%. It is estimated that about one third of children have a ‘definite’ orthodontic treatment need. A report by Allwright and Burndred32 provided the first published study of the prevalence of dentofacial anomalies requiring treatment intervention among Hong Kong children. Their study included 31% of the 6-11-year-old children who participated in the 1962 oral health survey6 and reported that 40.9% of the children exhibited certain dentofacial anomalies. The most common malocclusions were crowding (20.3%), maxillary overjet (14.5%), mandibular overjet (8.1%), overbite (6.9%), spacing (2.9%), and open bite (1.1%). The prevalence of handicapping dentofacial anomalies was higher among those aged 9 to 11 years (54.2%) than among those aged 6 to 8 years (36.4%).

Wong7 reported that approximately one (20.6%) in five of 5-14-year-old children had dentofacial anomalies that required dental treatment. The most common dentofacial anomaly was crowding (17.6%), followed by maxillary overjet (9.5%), overbite (4.2%), mandibular overjet (2.7%), spacing (1.1%), and open bite (0.7%). The prevalence of anomalies was higher among those aged 9 to 15 years (28.2%) than among those aged 5 to 8 years (16.4%). The reduction in the prevalence estimates of malocclusion when compared with the 1960 study6 could be due to the participation of different examiners (orthodontists in 1960 study vs general dental practitioners in 1968 study) and the wider age range of children involved in the 1968 survey.7

In the 1980s, the reported prevalence of dentofacial anomalies that required treatment was 10% to 36%.8 10 11 The percentage of children with cleft palates and/or lip was approximately 0.2%, with a slightly higher proportion of children with cleft lip.8 11 The most common reported dentofacial anomalies that required treatment were crowding (3.3-18.5%), maxillary overjet (2.8-5.1%), cross-bite (2.4-7.5%), reverse overjet anomaly (1%), deep overbite (0.9-5.4%), and open bite (0.5-1.9%). The prevalence of dentofacial anomalies was higher among 9-11-year-old children (13-38%) compared with 6-8-year-old children (6.7-34%).8 11 Among all the reported studies, only Wang et al33 used DHC IOTN (Dental Health Component of the Index of Orthodontic Treatment Need)34 to assess treatment need. The prevalence of malocclusion was estimated to be 88% and over a third (37%) of the study sample were deemed to have a ‘definite’ orthodontic treatment need and 33% had a moderate need.

The introduction of water fluoridation in the 1960s resulted in improvements in dental caries in Hong Kong. Prior to its implementation, nearly all children in the population had dental caries, with a high mean number of decayed, missing, or filled teeth because of tooth decay (mean dmft of 9.2 and DMFT of 4.4).5 The implementation of water fluoridation in the community led to a gradual decline in the caries experience and severity in children, as confirmed by the first three oral epidemiological studies in Hong Kong.5 6 7 The caries experience remained constant (plateaued) with approximately 50% of preschool children and 20% to 40% of schoolchildren since 1980s/1990s having a dental caries experience. This indicates that children in Hong Kong are benefiting from the continual effects of water fluoridation as well as exposure to fluoride from other sources, together with changing living conditions, lifestyles, and improved oral self-care habits in recent decades. The stable caries prevalence in recent years signifies that dental caries in children is controlled to a certain degree, but still remains prevalent, however.

Despite great improvements in the oral health of Hong Kong children over the past 50 years, dental caries remains an oral health burden in the community, in particular among preschool children where prevalence and incidence remain high. Although caries prevalence and severity among preschool children declined during the first 30 years following water fluoridation, the prevalence remains similar and its extent/severity has been even higher in recent decades.16 17 19 20 This suggests that there remain some preschool children for whom the current measures alone (water fluoridation and oral health education) are insufficient to ensure optimal oral health.

The dental caries prevalence and severity in preschool children tended to rise with increasing age.4 19 20 From the epidemiological studies, it is thus common to find a higher percentage of caries experience in children at the upper end of the preschool age range. Moreover, caries experience is not uniformly distributed within populations of children. Children from disadvantaged and socially marginalised populations had a higher caries experience and severity.4 18 19 20 Preventive measures and oral health education should start earlier among younger children and their parents or caregivers. In particular, efforts should focus on the underprivileged population in our community.

The dental caries condition among Hong Kong schoolchildren (12 years old) is relatively good by international comparisons. Dental caries affects 60% to 90% of schoolchildren in most industrialised countries.35 36 The current dental caries experience in 12-year-old Hong Kong children is relatively very low.16 17 Of note, most of the dental caries experience was related to filled teeth, few (approximately 1 in 20) had untreated decay. This pattern can be largely attributed to the contribution of the SDCS, which began in the 1980s as a school-based dental care system that effectively overcomes many social barriers to dental care access by schoolchildren (eg family income, education, dental health awareness). Schoolchildren receive regular quality dental care and treatment through the SDCS. The observed low level of untreated dental caries among schoolchildren is in stark contrast to findings prior to the introduction of SDCS when most decay remained untreated or was treated by extraction.5 6 7 8 The SDCS has also raised awareness of oral health among schoolchildren. Education about the importance of oral health has likely changed children’s lifestyle and improved their self-care practice and use of fluoride oral health care products.

Dental attendance among primary schoolchildren is high because of high participation in the SDCS,37 but remains worryingly low among secondary school and preschool children. Many adolescents and their parents do not consider there to be a need for such care. Less than a third of such children reported regular attendance for dental check-ups, presumably because this group of children have to be seen privately to access dental care.10 13 16 17 18 19 20 Early and regular dental check-ups to enable preventive care should be advocated.

The introduction of a number of public health measures in Hong Kong, mainly water fluoridation and the SDCS, has improved the oral health of Hong Kong children over the past 50 years. There has been a decline in dental caries among schoolchildren and adolescents. Nonetheless, the dental caries experience has remained unchanged in recent decades for preschool children; even a slight increase in extent/severity has been observed. Although there is evidence of improvement, the overall periodontal health of Hong Kong children remains unsatisfactory. A decrease in the prevalence and severity of enamel defects among Hong Kong children was observed, but there has recently been a slight increase. In view of the limited data regarding malocclusion in Hong Kong children, epidemiological studies should be considered. The utilisation of dental services is low, especially among preschool children who are not covered by the SDCS. New policies to develop dental care protocols to ensure evidence-based standards of care, and to advocate regular access to dental care and preventive services may further improve the oral health of Hong Kong children.

1. Evans RW, Lo EC, Lind OP. Changes in dental health in Hong Kong after 25 years of water fluoridation. Community Dent Health 1987;4:383-94.

2. Chiu GK, Davies WI. The historical development of dentistry in Hong Kong. Hong Kong Med J 1998;4:73-6.

3. Hong Kong Department of Health. Health Statistics. Surveillance &amp Epidemiology Branch, Centre for Health Protection. Health facts of Hong Kong. Available from: http://www.dh.gov.hk/english/statistics/statistics_hs/files/Health_Statistics_pamphlet_E.pdf. Accessed Oct 2014.

4. Chu CH, Fung DS, Lo EC. Dental caries status of preschool children in Hong Kong. Br Dent J 1999;187:616-20. Crossref

5. Medical and Health Department. Report on the 1st (pre-fluoridation) dental survey of school children in Hong Kong. Medical and Health Department, Hong Kong; 1960.

6. Medical and Health Department. Report on the 2nd (post-fluoridation) dental survey of school children in Hong Kong. Medical and Health Department, Hong Kong; 1962.

7. Wong KK. Report of a dental survey in Hong Kong 1968. The Government Dental Service, Hong Kong and the World Health Organization; 1968.

8. Law YH. Dental health status of primary school children in Hong Kong. The Bulletin. Hong Kong Soc Community Med 1981;12:1-16.

9. King NM, Ling JY, Ng BV, Wei SH. The dental caries status and dental treatment patterns of 12-year-old children in Hong Kong. J Dent Res 1986;65:1371-4. Crossref

10. Lind OP, Evans RW, Holmgren CJ, Corbet EF, Lim LP, Davies WI. Hong Kong Survey of Adult Oral Health 1984. Hong Kong: Department of Periodontology and Public Health, Faculty of Dentistry, University of Hong Kong; 1986.

11. Wong PY. A report on a dental survey of primary school children in Hong Kong. Medical and Health Department, Hong Kong; 1987.

12. Lo EC, Evans RW, Lind OP. Dental caries status and treatment needs of the permanent dentition of 6-12-year-olds in Hong Kong. Community Dent Oral Epidemiol 1990;18:9-11. Crossref

13. Kwan EL. Oral health status of 13 and 15 year-old secondary school children in Hong Kong [dissertation]. Hong Kong: University of Hong Kong; 1992.

14. Yiu C, Wong MC, Chan M, et al. Oral health of 15-year-old secondary school students in H.K. Hong Kong: Faculty of Dentistry, University of Hong Kong; 2001.

15. Evans RW, Lo EC. Effects of School Dental Care Service in Hong Kong—primary teeth. Community Dent Oral Epidemiol 1992;20:193-5. Crossref

16. Hong Kong Department of Health. Oral health survey 2001: common dental diseases and oral health related behaviour. Hong Kong SAR: Dental Services Head Office, Department of Health; 2002.

17. Hong Kong Department of Health. Oral health survey 2011: common dental diseases and oral health related behaviour. Hong Kong SAR: Dental Services Head Office, Department of Health; 2012.

18. Wei SH, Holm AK, Tong LS, Yuen SW. Dental caries prevalence and related factors in 5-year-old children in Hong Kong. Pediatr Dent 1993;15:116-9.

19. Lo EC, Loo EK, Lee CK. Dental health status of Hong Kong preschool children. Hong Kong Dent J 2009;6:6-12.

20. Chu CH, Ho PL, Lo EC. Oral health status and behaviours of preschool children in Hong Kong. BMC Public Health 2012;12:767. Crossref

21. Ainamo J, Barmes D, Beagrie G, Cutress T, Martin J, Sardo-Infirri J. Development of the World Health Organization (WHO) community periodontal index of treatment needs (CPITN). Int Dent J 1982;32:281-91.

22. Wu D. Provision of outreach dental service and caries risk assessment to preschool children in Hong Kong [dissertation]. Hong Kong: University of Hong Kong; 2011.

23. Ainamo J, Bay I. Problems and proposals for recording gingivitis and plaque. Int Dent J 1975;25:229-35.

24. King NM. Developmental defects of enamel in Chinese girls and boys in Hong Kong. Adv Dent Res 1989;3:120-5.

25. Evans RW. Changes in dental fluorosis following an adjustment to the fluoride concentration of Hong Kong’s water supplies. Adv Dent Res 1989;3:154-60.

26. Evans RW, Stamm JW. Dental fluorosis following downward adjustment of fluoride in drinking water. J Public Health Dent 1991;51:91-8. Crossref

27. Wong HM, McGrath C, King NM. Diffuse opacities in 12-year-old Hong Kong children—four cross-sectional surveys. Community Dent Oral Epidemiol 2014;42:61-9. Crossref

28. Wong HM, McGrath C, Lo EC, King NM. Association between developmental defects of enamel and different concentrations of fluoride in the public water supply. Caries Res 2006;40:481-6. Crossref

29. Wong HM, Peng SM, Wen YF, King NM, McGrath CP. Risk factors of developmental defects of enamel—a prospective cohort study. PLoS One 2014;9:e109351. Crossref

30. A review of the developmental defects of enamel index (DDE Index). Commission on Oral Health, Research & Epidemiology. Report of an FDI Working Group. Int Dent J 1992;42:411-26.

31. Cho SY, Ki Y, Chu V. Molar incisor hypomineralization in Hong Kong Chinese children. Int J Paediatr Dent 2008;18:348-52. Crossref

32. Allwright WC, Burndred WH. A survey of handicapping dentofacial anomalies among Chinese in Hong Kong. Int Dent J 1964;14:505-19.

33. Wang G, Hägg U, Ling J. The orthodontic treatment need and demand of Hong Kong Chinese children. Chin J Dent Res 1999;2:84-92.

34. Brook PH, Shaw WC. The development of an index of orthodontic treatment priority. Eur J Orthod 1989;11:309-20.

35. Petersen PE, Bourgeois D, Ogawa H, Estupinan-Day S, Ndiaye C. The global burden of oral diseases and risks to oral health. Bull World Health Organ 2005;83:661-9.

36. Dye BA, Tan S, Smith V, et al. Trends in oral health status: United States, 1988-1994 and 1999-2004. Vital Health Stat 11 2007;(248):1-92.

37. King NM. Paediatric dentistry in Hong Kong. Int J Paediatr Dent 1998;8:1-2. Crossref

Young boys are often brought by parents to see a medical practitioner for ‘phimosis’, and circumcision is one of the most commonly performed operations. Yet this topic is often not taught routinely in medical school. Buried penis is another less well-defined condition. In this review article, we will describe these conditions in a more systematic manner and present the current available knowledge about the conditions and management options.

Phimosis generally refers to a condition where the prepuce cannot be withdrawn to expose the glans. True phimosis, however, should be defined as a pathological condition in which the prepuce is scarred, non-retractile, and with a narrow preputial ring. This is secondary to balanitis xerotica obliterans (BXO). To avoid confusion of the terms, ‘physiological phimosis’ and ‘pathological phimosis’ should be used.

Physiological phimosis is a natural condition in which the prepuce cannot be retracted and there is natural adhesion between the glans and the prepuce (Fig 1). Almost all normal male babies are born with a non-retractable foreskin. Indeed, Gairdner1 noticed only 4% of newborns in England and Wales had retractable foreskin. The foreskin becomes retractable as the child grows. The adhesion between the prepuce and glans will also separate gradually as a spontaneous biological process.2 3 4 By the age of 3 years, 90% of prepuces are retractable.1 Øster2 examined preputial development in 173 Danish boys aged 6 to 17 years annually for 7 years and determined that the foreskin was non-retractable in 8% of young boys but in only 1% at 17 years of age. Similar findings were noted in Chinese boys by Ko et al5 and Hsieh et al,6 who reported 84.1%5 and 58.1%6 of boys with a completely retractable prepuce by the age of 13 years. Both the retractability and the shape of the prepuce lie within a spectrum that can sometimes be difficult to describe and there is no agreed classification system. Different papers have used their own classifications for the purpose of study. One example was the study by Kayaba et al3 in which retractability was graded according to how much of the glans was visible after prepuce retraction.

Balanitis xerotica obliterans is a chronic and progressive inflammatory condition that affects the prepuce, glans, and sometimes the urethra (Fig 2). It was first described in 1928 by Stühmer.7 There are three components of this condition: ‘balanitis’, meaning chronic inflammation of the glans penis; ‘xerotica’, an abnormally dry appearance of the lesion; and ‘obliterans’, for the association of occasional endarteritis.

The aetiology and the true incidence is unknown. An incidence of 0.6% has been reported for boys affected by their 15th birthday.8 It is suspected clinically when there is a ring of hardened tissue with a whitish colour at the tip of the foreskin. There are also other clinical features such as white patches over the glans, perimeatal sclerotic changes, or meatal stenosis. It can cause urethral stricture and retention of urine.

Medical students are not taught about the condition and it is generally not diagnosed at the primary care level. Gargollo et al9 reviewed 41 patients with the pathologically confirmed diagnosis of BXO at their centre and confirmed that no patient had the diagnosis at referral. Pathology of the excised prepuce showed lymphocytic infiltration in the ripper dermis, hyalinosis and homogenisation of collagen, basal cell vacuolation, atrophy of the stratum malpighii, and hyperkeratosis.

Parents often seek medical advice about their son’s ‘foreskin problems’. Pain, redness, itchiness, long prepuce, ballooning during urination, difficulty in retracting the prepuce, and penis being too short are the common complaints. Before answering all the questions, we should be able to differentiate the normal and abnormal.

Most parents may think that the presence of pain or pruritus indicates infection of the prepuce, and yet poor prepuce hygiene is a more common problem. Smegma is another common complaint from parents, usually described as a ‘mass’, or ‘white pearl’. Smegma can be identified by gently retracting the prepuce (Fig 3). It is harmless and is a combination of secretions and desquamated skin.

Difficulty in retracting the prepuce and long prepuce is a feature of physiological phimosis. This is normal in most boys and requires no attention apart from daily routine prepuce hygiene. The role of the physician is to differentiate normal and abnormal prepuce, then guide proper management.

Ballooning is a feature of a tight prepuce. Because of the tight preputial opening, there is dilatation of the preputial sac during voiding. This causes a lot of parental anxiety about possible urinary outflow obstruction. Babu et al10 performed uroflow studies in boys with and without ballooning of the foreskin and determined that there was significant difference.

This refers to inflammation of the glans (balanitis) and the foreskin (posthitis) [Fig 4]. Patients present with a swollen prepuce with or without discharge from the preputial opening. It is a relatively common condition, with a reported incidence of 6% in uncircumcised boys.11 In the absence of fever, underlying urinary tract infection (UTI) is unlikely. Simple bathing and rinsing with normal saline or chlorhexidine gluconate solution after urination is sufficient treatment for afebrile patients. Topical antibiotic cream is commonly prescribed for local infection. Serious conditions and presence of fever may warrant further investigations, oral antibiotics, or even hospital admission.

After diagnosing physiological phimosis, parents should be taught how to keep the prepuce clean. Only a small proportion of parents know what is required.12 Gentle daily retraction of the prepuce and rinsing of the prepuce with warm water can maintain good hygiene and prevent infection. Parents should also be taught to avoid forcible retraction of the prepuce.13 14 15 Simple stretching of the prepuce alone has been shown to be effective in achieving complete resolution of physiological phimosis.16 After 3 months of prepuce stretching, 76% of patients reported resolution of phimosis.16

Topical steroids have been prescribed in the treatment of phimosis. Their anti-inflammatory, immunosuppressive, and skin-thinning properties are believed to be the mechanism for resolution of phimosis.17 Their use in physiological phimosis was first described by Kikiros et al.18 Subsequent studies showed the response rate for resolution of phimosis to be 68.2% to 95%.16 19 20 21 22 Moreno et al23 subsequently performed a meta-analysis and reviewed 12 randomised controlled trials on the use of different topical steroid formulations, and again confirmed the significant benefit of corticosteroids in the complete or partial clinical resolution of phimosis (risk ratio=2.45; 95% confidence interval, 1.84-3.26).

Parents often ask about the potential complications of topical steroid use. Golubovic et al24 and Pileggi et al25 addressed this issue by measuring serum cortisol levels and salivary cortisol levels, respectively. Neither could demonstrate a significant change in cortisol level after application of topical steroids. Topical steroid therapy is thus a safe and effective alternative to circumcision.

Circumcision is a procedure in which part of the foreskin is removed and results in a non-covered glans. It is a procedure that has been described for many years and is performed almost universally in Jewish and Muslim boys. The rate of newborn circumcision is high in the US (>50%),26 27 but routine circumcision is not a tradition in the Chinese population. Leung et al28 showed the circumcision rate in 6- to 12-year-old boys in Hong Kong to be 10.7%.

There is evidence that circumcision can reduce the risk of UTI, penile cancer, human immunodeficiency virus (HIV), and sexually transmitted disease (STD).

Childhood UTI is associated with renal scarring. The symptoms and signs of UTI are often non-specific in young children who may present with fever alone. The overall prevalence of UTI in children with fever (<19 years old) was reported to be 7.8% in a meta-analysis published in 2008.29 The pooled prevalence of febrile UTI in male infants from 0 to 24 months of age was 8.0% (confidence interval, 5.5-10.4%). Circumcised boys had a lower risk of developing UTI—20.1% in uncircumcised versus 2.4% in circumcised infants of less than 3 months of age with fever.29

Another systematic review in 2005 showed a decreased risk of UTI in circumcised boys.30 The authors calculated the number-needed-to-treat was 111 in normal boys, but the number-needed-to-treat for recurrent UTI and high-grade vesicoureteric reflux was 11 and 4, respectively. It was evident that the benefits of circumcision were higher for boys at risk of UTI.

Three randomised controlled trials concluded that adult circumcision had a protective effect against acquisition of HIV.31 32 33 Although the full mechanism of protection was not fully understood, it was shown that the inner foreskin harbours epithelial CD4+ CCR5+ cells and has features of an inflamed epidermal barrier. These changes may support a subclinical inflammatory state in uncircumcised men, with availability of target cells for HIV infection, and potentially account for the benefits of circumcision in STD prevention.34 All the trials were performed in Africa, with a much higher prevalence of HIV. Education about use of condoms and safe sex practice was relatively primitive compared with Hong Kong. Readers should therefore interpret these results with caution when discussing the benefits of circumcision on HIV prevention with our patients.

A meta-analysis published in 2006 by Weiss et al35 showed that circumcised males were at a lower risk of syphilis, and there was a lower association with herpes simplex virus (HSV) type 2. Other cohorts also showed similar findings, with circumcised males having a decreased risk of syphilis, gonorrhoea, and human papillomavirus (HPV).36 37 38 39 On the contrary, another systematic review by Van Howe40 showed that most STDs are not impacted significantly by circumcision status. They included chlamydia, gonorrhoea, HSV, and HPV. Despite the positive findings in some studies, it should be remembered that the use of a condom and safe sex are the most important deterrents. The protective effect of circumcision might give a false sense of security and should not be advocated over other preventive measures.

Circumcision is one of the most commonly performed operations in the world and involves excision of a ring of preputial tissue. In general, the procedure may involve the use of a special device (eg Plastibell, Gomco clamp, Mogen clamp, Shang ring) or may apply the ‘free hand excision method’. Depending on the method, suturing may or may not be involved. Every procedure is associated with risks and complications. The rate is different depending on the operator (ritual circumciser or surgeon) and the setting (home, clinic, or hospital).

There has been inadequate comparison of the complication rates of ‘device method’ surgery and the ‘free hand excision method’ so it is difficult for the authors to recommend a single best method for surgical circumcision. The overall complication rate of circumcision varies from 0.5%41 to 8%.42 As the indication for circumcision in some patients is not medical (eg religious or ritual circumcision), the risks should be carefully explained to the patient before the procedure.

Early complications include bleeding, wound infection, and UTI. Bleeding is one of the most common postoperative complications that, in extreme cases, may lead to shock.43 Meticulous technique during the procedure is thus important. If bleeding is encountered postoperatively, it can usually be controlled by local compression or bedside plication.

In a UK study, the infection rate after circumcision has been reported to be around 0.3%.44 It is usually minor and can be treated by simple irrigation with antiseptic solution. Systemic antibiotics are rarely needed. A systematic review of the prevalence and complications of circumcision was performed in eastern and southern Africa.45 The infection rate was very high and two thirds of patients presented with systemic infection requiring antibiotics. The authors believe the quality of local wound care is very important in minimising the infection rate. Urinary retention is uncommon after circumcision but can occur in up to 3.6% of cases. It is likely to be pain-related or due to improper placement of a circumcision device, eg Plastibell.46

Wound dehiscence may occur and can be managed with wound care and dressing. Very rarely, excessive prepuce loss as a result of excessive skin excision may be seen. This potentially disastrous complication has been reported to be treated with a full-thickness skin graft.47

Late complications are not uncommon, reported by one study to be present in 4.7% of newborn circumcisions.48 Redundant residual skin and recurrent penile adhesion are the two most common late complications that may necessitate revision circumcision. Meatal stenosis is another late but uncommon complication after circumcision and requires surgery. The cause is not known but it is more common in patients with BXO.49 50

There are some other less-common but severe complications, including urethrocutaneous fistula, glans amputation, and iatrogenic buried penis.51 52 These have long-term physiological and psychological consequences for both the patient and family. Surgical technique and the surgeon’s awareness of limitations of each method of circumcision is important. Parents should be fully informed before they make a decision about circumcision, especially where the patient is physically weak or where there is no medical indication for the procedure.

Various international colleges have produced guidelines on circumcision. These include the British Association of Paediatric Surgeons,53 the Royal Australasian College of Physicians,54 and the American Academy of Pediatrics.55 56

Having taken all these into account, the overall view of our unit is as follows:
(1) Although there is some scientific evidence for the benefits of circumcision, the routine use on all males is not justified. Parents should be fully informed of all the potential benefits and risks of the procedure.
(2) Our current medical indications for circumcision are:
(a) penile malignancy (though this is extremely rare in children) or traumatic foreskin injury where it cannot be salvaged; and
(b) BXO, severe recurrent attacks of balanoposthitis, and/or recurrent febrile UTIs.
(3) Non-therapeutic ‘ritual’ circumcision may be offered.

Buried penis is a condition where the penis is ‘trapped’ or ‘concealed’ under the suprapubic area. There is an apparent absence or partial absence of the penis. Figure 5 shows partial buried penis in an 8-year-old boy. The condition was described as ‘complete’ or ‘partial’ by Crawford57 in 1977. In the partial type, the proximal half of the penile shaft is buried in the subcutaneous tissue. For the complete type, the phallus is completely invisible and the glans is covered only by prepuce. Maizels et al58 further elaborated in 1986, offering new classifications as ‘buried penis’ (patients with redundant suprapubic fat and/or lack of penile skin anchoring to deep fascia), ‘webbed penis’ (scrotal skin webs the penoscrotal angle to obscure the penis), ‘trapped penis’ (the shaft of the penis is entrapped in the scarred, prepubic skin following trauma/overzealous circumcision), ‘micro-penis’ (a normally formed penis that is less than two standard deviations below the mean size in the stretched length), and ‘diminutive penis’ (a penis that is small and/or malformed as a consequence of epispadias/exstrophy, severe hypospadias, disorder of sexual differentiation, or chromosomal anomalies). O’Brien et al59 described another condition called ‘congenital megaprepuce’ in 1994 that includes a phimotic ring and large preputial sac. Despite these studies, buried penis is still not a well-defined or well-classified entity. It can be congenital or iatrogenic after overzealous circumcision. Clinicians are reminded to examine the penis carefully and the exact penile length should be a properly performed ‘stretched penile length’. When there is uncertainty about the exact diagnosis, early specialist advice is advocated.

For congenital problems, anxious parents usually seek medical advice because they feel that their child’s penis is too short. Other problems include local infection, urinary retention, inability to void standing, chronic urinary dripping, and undirected voiding. For older children, there may be pain during erection or disturbed vaginal penetration.60 61 62

Anatomically, buried penis is usually due to insufficient outer prepuce and lack of attachment between the penile Buck’s fascia and the pubis.63 Numerous corrective surgical techniques have been described. The underlying principle is the degloving of the penis, anchoring of Buck’s fascia to the pubis, and preputioplasty (pedicled preputial flap, Z-plasty of the prepuce, lipectomy, and skin graft) [Fig 6].64 65

A study on the comparison of quality of life before and after surgery showed significant improvement in sexual pleasure, urination difficulties, and genital hygiene.66 King et al61 also reported that all patients were happy with the aesthetic results.

It is essential to recognise the features of physiological versus pathological phimosis. Physiological phimosis (tightness of prepuce), preputial adhesion, and smegma are common and normal in young boys, and do not require surgical intervention. There are potential benefits and complications of circumcision that should be thoroughly appreciated by physicians before discussion with parents or patient. Medical indications for circumcision include penile malignancy, traumatic foreskin injury, recurrent attacks of severe balanoposthitis, and recurrent febrile UTIs with abnormal urinary tract. Very few international societies support routine circumcision despite the potential medical benefits incurred. Buried penis is a condition that may warrant surgery.

1. Gairdner D. The fate of the foreskin, a study of circumcision. Br Med J 1949;2:1433-7. Crossref

2. Øster J. Further fate of the foreskin. Incidence of preputial adhesions, phimosis, and smegma among Danish schoolboys. Arch Dis Child 1968;43:200-3. Crossref

3. Kayaba H, Tamura H, Kitajima S, Fujiwara Y, Kato T, Kato T. Analysis of shape and retractability of the prepuce in 603 Japanese boys. J Urol 1996;156:1813-5. Crossref

4. Rickwood AM. Medical indications for circumcision. BJU Int 1999;83 Suppl 1:45-51. Crossref

5. Ko MC, Liu CK, Lee WK, Jeng HS, Chiang HS, Li CY. Age-specific prevalence rates of phimosis and circumcision in Taiwanese boys. J Formos Med Assoc 2007;106:302-7. Crossref

6. Hsieh TF, Chang CH, Chang SS. Foreskin development before adolescence in 2149 schoolboys. Int J Urol 2006;13:968-70. Crossref

7. Stühmer A. Balanitis xerotica obliterans (post operationem) und ihre Beziehungenzur ‘Kraurosis glandi et praeputii penis’ [in German]. Arch Dermatol Syph (Berlin) 1928;156:613. Crossref

8. Shankar KR, Rickwood AM. The incidence of phimosis in boys. BJU Int 1999;84:101-2. Crossref

9. Gargollo PC, Kozakewich HP, Bauer SB, et al. Balanitis xerotica obliterans in boys. J Urol 2005;174:1409-12. Crossref

10. Babu R, Harrison SK, Hutton KA. Ballooning of the foreskin and physiological phimosis: is there any objective evidence of obstructed voiding? BJU Int 2004;94:384-7. Crossref

11. Herzog LW, Alvarez SR. The frequency of foreskin problems in uncircumcised children. Am J Dis Child 1986;140:254-6. Crossref

12. Osborn LM, Metcalf TJ, Mariani EM. Hygienic care in uncircumcised infants. Pediatrics 1981;67:365-7.

13. Camille CJ, Kuo RL, Wiener JS. Caring for the uncircumcised penis: what parents (and you) need to know. Contemp Pediatr 2002;11:61.

14. Simpson ET, Barraclough P. The management of the paediatric foreskin. Aust Fam Physician 1998;27:381-3.

15. Rickwood AM, Hemalatha V, Batcup G, Spitz L. Phimosis in boys. Br J Urol 1980;52:147-50. Crossref

16. Zampieri N, Corroppolo M, Camoglio FS, Giacomello L, Ottolenghi A. Phimosis: stretching methods with or without application of topical steroids? J Pediatr 2005;147:705-6. Crossref

17. Kragballe K. Topical corticosteroids: mechanisms of action. Acta Derm Venereol Suppl (Stockh) 1989;151:7-10; discussion 47-52.

18. Kikiros CS, Beasley SW, Woodward AA. The response of phimosis to local steroid application. Pediatr Surg Int 1993;8:329-32. Crossref

19. Lee CH, Lee SD. Effect of topical steroid (0.05% clobetasol propionate) treatment in children with severe phimosis. Korean J Urol 2013;54:624-30. Crossref

20. Chu CC, Chen KC, Diau GY. Topical steroid treatment of phimosis in boys. J Urol 1999;162:861-3. Crossref

21. Orsola A, Caffaratti J, Garat JM. Conservative treatment of phimosis in children using a topical steroid. Urology 2000;56:307-10. Crossref

22. Ghysel C, Vander Eeckt K, Bogaert GA. Long-term efficiency of skin stretching and a topical corticoid cream application for unretractable foreskin and phimosis in prepubertal boys. Urol Int 2009;82:81-8. Crossref

23. Moreno G, Corbalán J, Peñaloza B, Pantoja T. Topical corticosteroids for treating phimosis in boys. Cochrane Database Syst Rev 2014;(9):CD008973. Crossref

24. Golubovic Z, Milanovic D, Vukadinovic V, Rakic I, Perovic S. The conservative treatment of phimosis in boys. Br J Urol 1996;78:786-8. Crossref

25. Pileggi FO, Martinelli CE Jr, Tazima MF, Daneluzzi JC, Vicente YA. Is suppression of hypothalamic-pituitary-adrenal axis significant during clinical treatment of phimosis? J Urol 2010;183:2327-31. Crossref

26. Centers for Disease Control and Prevention (CDC). Trends in in-hospital newborn male circumcision—United States, 1999-2010. MMWR Morb Mortal Wkly Rep 2011;60:1167-8.

27. Warner L, Cox S, Kuklina E, et al. Updated trends in the incidence of circumcision among male newborn delivery hospitalizations in the United States, 2000-2008. Proceedings of the National HIV Prevention Conference; 2011 Aug 26; Atlanta, Georgia, US.

28. Leung MW, Tang PM, Chao NS, Liu KK. Hong Kong Chinese parents’ attitudes towards circumcision. Hong Kong Med J 2012;18:496-501.

29. Shaikh N, Morone NE, Bost JE, Farrell MH. Prevalence of urinary tract infection in childhood: a meta-analysis. Pediatr Infect Dis J 2008;27:302-8. Crossref

30. Singh-Grewal D, Macdessi J, Craig J. Circumcision for the prevention of urinary tract infection in boys: a systematic review of randomised trials and observational studies. Arch Dis Child 2005;90:853-8. Crossref

31. Gray RH, Kigozi G, Serwadda D, et al. Male circumcision for HIV prevention in men in Rakai, Uganda: a randomised trial. Lancet 2007;369:657-66. Crossref

32. Bailey RC, Moses S, Parker CB, et al. Male circumcision for HIV prevention in young men in Kisumu, Kenya: a randomised controlled trial. Lancet 2007;369:643-56. Crossref

33. Auvert B, Taljaard D, Lagarde E, Sobngwi-Tambekou J, Sitta R, Puren A. Randomized, controlled intervention trial of male circumcision for reduction of HIV infection risk: the ANRS 1265 Trial. PLoS Med 2005;2:e298. Crossref

34. Lemos MP, Lama JR, Karuna ST, et al. The inner foreskin of healthy males at risk of HIV infection harbors epithelial CD4+ CCR5+ cells and has features of an inflamed epidermal barrier. PLoS One 2014;9:e108954. Crossref

35. Weiss HA, Thomas SL, Munabi SK, Hayes RJ. Male circumcision and risk of syphilis, chancroid, and genital herpes: a systematic review and meta-analysis. Sex Transm Infect 2006;82:101-9; discussion 110. Crossref

36. Diseker RA 3rd, Peterman TA, Kamb ML, et al. Circumcision and STD in the United States: cross sectional and cohort analyses. Sex Transm Infect 2000;76:474-9. Crossref

37. Albero G, Castellsagué X, Giuliano AR, Bosch FX. Male circumcision and genital human papillomavirus: a systematic review and meta-analysis. Sex Transm Dis 2012;39:104-13. Crossref

38. Tobian AA, Serwadda D, Quinn TC, et al. Male circumcision for the prevention of HSV-2 and HPV infections and syphilis. N Engl J Med 2009;360:1298-309. Crossref

39. Turner AN, Morrison CS, Padian NS, et al. Male circumcision and women’s risk of incident chlamydial, gonococcal, and trichomonal infections. Sex Transm Dis 2008;35:689-95. Crossref

40. Van Howe RS. Sexually transmitted infections and male circumcision: a systematic review and meta-analysis. ISRN Urol 2013;2013:109846. Crossref

41. El Bcheraoui C, Zhang X, Cooper CS, Rose CE, Kilmarx PH, Chen RT. Rates of adverse events associated with male circumcision in U.S. medical settings, 2001 to 2010. JAMA Pediatr 2014;168:625-34. Crossref

42. Weiss HA, Larke N, Halperin D, Schenker I. Complications of circumcision in male neonates, infants and children: a systematic review. BMC Urol 2010;10:2. Crossref

43. Bocquet N, Chappuy H, Lortat-Jacob S, Chéron G. Bleeding complications after ritual circumcision: about six children. Eur J Pediatr 2010;169:359-62. Crossref

44. Cathcart P, Nuttall M, van der Meulen J, Emberton M, Kenny SE. Trends in paediatric circumcision and its complications in England between 1997 and 2003. Br J Surg 2006;93:885-90. Crossref

45. Wilcken A, Keil T, Dick B. Traditional male circumcision in eastern and southern Africa: a systematic review of prevalence and complications. Bull World Health Organ 2010;88:907-14. Crossref

46. Mihssin N, Moorthy K, Houghton PW. Retention of urine: an unusual complication of the Plastibell device. BJU Int 1999;84:745. Crossref

47. Özdemir E. Significantly increased complication risks with mass circumcisions. Br J Urol 1997;80:136-9. Crossref

48. Patel HI, Moriarty KP, Brisson PA, Feins NR. Genitourinary injuries in the newborn. J Pediatr Surg 2001;36:235-9. Crossref

49. Pieretti RV, Goldenstein AM, Pieretti-Vanmarcke R. Late complications of newborn circumcision: a common and avoidable problem. Pediatr Surg Int 2010;26:515-8. Crossref

50. Homer L, Buchanan KJ, Nasr B, Losty PD, Corbett HJ. Meatal stenosis in boys following circumcision for lichen sclerosus (balanitis xerotica obliterans). J Urol 2014;192:1784-8. Crossref

51. Ceylan K, Burhan K, Yilmaz Y, Can S, Kuş A, Mustafa G. Severe complications of circumcision: an analysis of 48 cases. J Pediatr Urol 2007;3:32-5. Crossref

52. Ince B, Gundeslioglu AO. A salvage operation for total penis amputation due to circumcision. Arch Plast Surg 2013;40:247-50. Crossref

53. Management of foreskin conditions. British Association of Paediatric Surgeons. Available from: http://www.baps.org.uk/resources/documents/management-foreskin-conditions/. Accessed Jun 2015.

54. The Royal Australasian College of Physicians. Circumcision of infant males. 2010. Available from: https://www.racp.edu.au/docs/default-source/advocacy-library/circumcision-of-infant-males.pdf. Accessed Jun 2015.

55. American Academy of Pediatrics Task Force on Circumcision. Circumcision policy statement. Pediatrics 2012;130:585-6. Crossref

56. American Academy of Pediatrics Task Force on Circumcision. Male circumcision. Pediatrics 2012;130:e756-85. Crossref

57. Crawford BS. Buried penis. Br J Plast Surg 1977;30:96-9. Crossref

58. Maizels M, Zaontz M, Donovan J, Bushnick PN, Firlit CF. Surgical correction of the buried penis: description of a classification system and a technique to correct the disorder. J Urol 1986;136:268-71.

59. O’Brien A, Shapiro AM, Frank JD. Phimosis or congenital megaprepuce? Br J Urol 1994;73:719-20. Crossref

60. Mattsson B, Vollmer C, Schwab C, et al. Complications of a buried penis in an extremely obese patient. Andrologia 2012;44 Suppl 1:826-8. Crossref

61. King IC, Tahir A, Ramanathan C, Siddiqui H. Buried penis: evaluation of outcomes in children and adults, modification of a unified treatment algorithm, and review of the literature. ISRN Urol 2013;2013:109349. Crossref

62. Donatucci CF, Ritter EF. Management of the buried penis in adults. J Urol 1998;159:420-4. Crossref

63. Chin TW, Tsai HL, Liu CS. Modified prepuce unfurling for buried penis: a report of 12 years of experience. Asian J Surg 2015;38:74-8. Crossref

64. Liu X, He DW, Hua Y, Zhang DY, Wei GH. Congenital completely buried penis in boys: anatomical basis and surgical technique. BJU Int 2013;112:271-5. Crossref

65. Chu CC, Chen YH, Diau GY, Loh IW, Chen KC. Preputial flaps to correct buried penis. Pediatr Surg Int 2007;23:1119-21. Crossref

66. Hughes DB, Perez E, Garcia RM, Aragón OR, Erdmann D. Sexual and overall quality of life improvements after surgical correction of “buried penis”. Ann Plast Surg 2016;76:532-5. Crossref

Glaucoma is a disease characterised by progressive optic nerve thinning. Intra-ocular pressure (IOP) remains an important modifiable risk factor even in normal- or low-tension glaucomas. Anti-glaucoma eye drops remain the mainstay of IOP-lowering therapy but their side-effects include conjunctival injection, allergy, hypertrichiasis, iris pigmentation, cystoid macular oedema, bradycardia, and bronchospasm. Laser trabeculoplasty describes the use of laser on the trabecular meshwork to promote aqueous output that in turn lowers the IOP. Laser trabeculoplasty is often performed when medications alone are inadequate for disease control, to avoid the side-effects of medication, or as a bridging therapy prior to more invasive glaucoma filtration surgery. Selective laser trabeculoplasty (SLT) was approved by the US Food and Drug Administration in 2001 for the treatment of open-angle glaucoma (OAG). Following the first clinical trial for the Hong Kong population in 2004 by Lai et al,1 the procedure has since gained popularity in our locality. Based on our understanding of the former argon laser trabeculoplasty technology, we are aware that variations in race and trabecular meshwork pigmentation can potentially influence laser success.2 3 4 This review aimed to summarise the published data on use of SLT in the Hong Kong Chinese population.

An Ovid search was made on 4 June 2015 using “selective laser trabeculoplasty” as the key word, which identified 190 unique articles; 24 reviews and/or meta-analyses were excluded. All remaining abstracts of original articles were in English. There were 15 abstracts in which an ophthalmology institution from Hong Kong Special Administrative Region (HKSAR) was identified as an affiliated institution. These 15 articles were reviewed to ensure that the study population contained participants from HKSAR prior to their inclusion in this review.

Selective laser trabeculoplasty is a low-energy, Q-switched, frequency-doubled Nd:YAG laser with a wavelength of 532 nm. Lasers are delivered at a fixed duration of 3 nanoseconds and spot size of 400 µm. The initial energy is 0.8 mJ, titrated up until bubble formation is just visible and titrated down if pain is experienced. The pigmented area is usually treated with confluent, non-overlapping laser spots in a 360° fashion in a single session.1 5 6 In cases of angle closure, SLT is applied to at least 90° to 180° of the visible trabecular meshwork, avoiding areas of peripheral anterior synechiae.7 8

An alpha-adrenergic agonist may be applied prior to or immediately after the procedure to prevent IOP spikes that can occur in up to 10% of subjects within the first 1 to 2 hours of treatment.1 Post-laser eye drops may vary from no eye drops, topical non-steroidal eye drops, or a weak topical steroid for a short duration. Excess inflammatory suppression should be avoided as SLT works via an inflammatory cascade of cytokine up-regulation, phagocytic activity, and trabecular matrix metalloproteinase expression that reduce outflow resistance at the trabecular meshwork.9 10 11 The majority of anterior chamber reactions following SLT are mild and spontaneous, and resolve within the first week of laser.12

Conventionally, 80 to 100 laser spots are delivered. Lee et al13 postulated that it is the total energy density (number of spots multiplied by the mean energy) delivered that is important, rather than just the number of laser spots. In a group of 49 Chinese OAG subjects who received 360° SLT treatment, bandwidth selection by generalised cross-validation methods was used to determine the optimal interval and point of total SLT energy that provided the greatest IOP reduction. The 95% confidence band by bootstrap analysis revealed that at energy intervals between 214.6 and 234.9 mJ, the IOP was most likely to decrease by 25%, with the optimal total energy at 226.1 mJ.

Lai et al1 conducted the first randomised controlled trial of SLT in a Hong Kong Chinese population with primary open-angle glaucoma (POAG) and ocular hypertension (OHT), comparing the effect of the laser versus topical anti-glaucoma medications alone over 5 years using the untreated eye as a control. Additional medical therapy was permitted for eyes that had inadequate IOP control despite SLT. In the study, 29 patients (17 POAG and 12 OHT) completed follow-up for 5 years and all were of Chinese ethnicity with a dark brown iris. The SLT eye had a mean IOP reduction of 32% (26.8 mm Hg to 18.3 mm Hg) at 5 years although this level of IOP reduction was statistically similar to the control eye. The SLT-treated eye, however, required fewer medications to maintain an IOP of ≤21 mm Hg (P<0.001); only 27.6% of eyes that received SLT required medications at 5 years. The failure rate, which was defined as an IOP of >21 mm Hg with maximal topical medications, was similar in both treatment arms (17.2% in the SLT-treated eye vs 27.6% in the eye that received medication alone; P=0.53).1

Similarly, in a randomised controlled trial studying the change in quality of life between subjects prescribed adjuvant SLT versus medication for the treatment of POAG, Lee et al5 reported that at 6 months, the SLT group (n=22) had a 7.6% lower IOP (P=0.03) and required 40.0% less medication (P=0.02) compared with the medical group (n=19). When compared with its own baseline at 6 months, the SLT group had a 15.1% lower IOP (P<0.0001) on top of a 34.8% reduction in medication requirement (P<0.0001) while the medical therapy group demonstrated no change in IOP or medication requirement (P>0.8). Despite these significant reductions in IOP and medications in the SLT group, however, there was no statistical difference between the translated Chinese version of the Glaucoma Quality of Life–15 score or the simplified Comparison of Ophthalmic Medications for Tolerability survey score between both treatment arms (P>0.2). The absence of a detectable change in the short-term quality of life could be related to the design of the existing glaucoma-specific quality-of-life surveys that mainly focus on detecting changes in vision, dark adaptation, and outdoor mobility, and may be suboptimal in assessing the side-effects and inconvenience of medication use.

Selective laser trabeculoplasty seems to be an effective alternative to medication with sustainable IOP reductions for up to 5 years together with a lower medication requirement.

Primary angle-closure glaucoma (PACG) is characterised by ≥270° of angle closure where the trabecular meshwork is on Shaffer grades 0 to 1. Ho et al7 reported the SLT outcome in 60 PACG patients with an IOP of >21 mm Hg in the presence of a patent laser iridotomy with >90° of visible pigmented trabecular meshwork. In the study, SLT was delivered to 90° of open angle and at 6 months, 82% of subjects had >3 mm Hg reduction, 72% had >4 mm Hg reduction, 54% had ≥20% IOP reduction, and 24% had ≥30% IOP reduction. The mean IOP decreased from 24.6 mm Hg to 18.7 mm Hg after SLT, signifying a 24% IOP reduction at 6 months.7 8

In a recent three-centre randomised clinical trial involving Singapore, Jakarta, and Hong Kong,8 100 subjects with PACG or primary angle closure (angle closure without glaucomatous optic neuropathy) with 180° of visible trabecular meshwork were randomised to receive SLT versus travoprost 0.004% for IOP control. At 6 months, both treatment arms had a similar degree of IOP reduction (16.9% in the SLT group vs 18.5% in the medical therapy group; P=0.52). The absolute success (IOP ≤21 mm Hg without medications) was 60.0% in the SLT group and 84.0% in the medical therapy group (P=0.008). Selective laser trabeculoplasty is effective in primary angle closure or PACG where at least 90° of the pigmented trabecular meshwork is accessible.

Normal-tension glaucoma (NTG) is a subtype of POAG with the presence of glaucomatous optic neuropathy but IOP never exceeds 21 mm Hg. There is a high prevalence in Korea and Japan, accounting for 77% and 92% of their POAG cases respectively, although there is no published prevalence for Hong Kong.14 15 The Collaborative Normal-Tension Glaucoma Study Group has demonstrated that a 30% IOP reduction may slow disease progression of NTG despite a non–pressure dependent, hypoperfusion element in the disease pathophysiology.16

Lee et al6 17 prospectively treated 46 patients with NTG who were currently prescribed topical anti-glaucoma medications. They received a 1-month washout of medication and a single session of SLT before medications were gradually resumed in a step-wise manner. Selective laser trabeculoplasty reduced IOP by 20% and 15% at 6 months and 1 year, respectively. Medications were also reduced by 27% at both 6 months and 1 year after laser.6 17 At 2 years, 34 subjects completed follow-up. The mean (± standard deviation) age was 65.1 ± 12.1 years with a post-washout IOP of 16.2 ± 2.3 mm Hg. The mean SLT shots delivered was 191.0 ± 27.3 with a mean energy of 1.0 ± 0.08 mJ. The mean IOP reduction at 24 months was 22% (P<0.0001) from the post-washout (without medication) level and 11.5% from the pre-study (with medication) level (P<0.0001), in addition to a medication reduction of 40% (P<0.0001).

There was a gradual decline in the absolute success rate (IOP reduction >20% without medication) following SLT, from 61% at 6 months, to 22% at 12 months, and to 11% at 24 months. It is well known that the effect of SLT decreases with time but the process may be repeated as needed.6 Although the effect of SLT on NTG is not as prominent as in those with POAG or PACG, it remains a worthwhile procedure in NTG where drug compliance may be an issue. Patients are often asymptomatic in the early stages and are told they have ‘normal’ pressures during follow-up, compromising their understanding of the disease. Nonetheless, it must be noted that SLT only addresses the pressure-dependent component of the disease and not the hypoperfusion element of NTG that can be equally important in disease progression.

The response or success rate to SLT is most commonly defined in the literature as an IOP reduction of ≥20%.18 19 This success rate varied among the different glaucoma subgroups in our local population however—47% in POAG,5 54% to 60% in PACG,7 8 and 60% in NTG.6 Likewise, the mean IOP reduction also varies among different glaucoma subgroups—32% reduction at 60 months in POAG,1 24% reduction at 6 months in PACG,7 and 22% reduction at 24 months for NTG.20

Not all treated eyes respond to SLT, thus understanding the factors that may influence a successful outcome is useful for both the clinician and patient. Lee et al21 22 23 analysed 25 potential covariates as detailed below, using univariate and multiple logistic regression analyses in Chinese subjects with OAG, NTG, and POAG independently. Success was defined as an IOP reduction of ≥20% while anti-glaucoma medications remained unchanged. The category of covariates included: type of glaucoma, age, gender, phakic status, presenting and pre-SLT IOP, IOP at various time intervals after SLT, number and type of anti-glaucoma medications, number of SLT spots, SLT energy, glaucoma severity via retinal nerve fibre layer thickness on optical coherence tomography and visual field index on Humphrey Visual Field, visual acuity, and pre-SLT corneal parameters.21 22 23

Among the 111 OAG eyes (60 NTG and 51 POAG) that were analysed, having a higher pre-laser IOP was a significant predictor of success on both univariate / multivariate analyses (coefficient=0.20 / 0.46, odds ratio [OR]=1.23 / 1.58, P=0.0017 / 0.0011). Use of three anti-glaucoma medications was more likely to result in a non-successful procedure on univariate / multivariate analyses (coefficient = –1.08 / –3.74, OR=0.3 / 0.024, P=0.037 / 0.0081).22

For the POAG subgroup, success was more likely in those with: an older age (coefficient=0.1, OR=1.1, P=0.0003), a higher pre-laser IOP (coefficient=0.3, OR=1.3, P=0.0005), four types of anti-glaucoma medications (coefficient=2.1, OR=8.4, P=0.005), a larger dioptre of spherical equivalent (coefficient=2.1, OR=8.4, P=0.005), and the use of a carbonic anhydrase inhibitor eye drop (coefficient=1.7, OR=6.0, P=0.003).21

In those with NTG, success was most commonly seen in those with a higher pre-laser IOP (coefficient=1.1, OR=3.1, P=0.05) and in those who achieved a lower IOP at 1 week after SLT (coefficient= –0.8, OR=0.5, P=0.04).23

Thus, it seems that regardless of the type of glaucoma, having a higher pre-laser IOP is one of the more consistent predictors of success. Nevertheless, it is important to keep in mind that the mean IOP reduction from SLT is around 22% to 32% as detailed in the earlier sections, and SLT alone may not adequately manage those with extremely high pressures.

Lai et al1 have reported that in our population, the incidence of IOP spikes of >5 mm Hg after SLT can occur in up to 10.3% of patients within the first 1 to 2 hours of SLT. Ho et al7 reported that 2.0% of their PACG population had IOP spikes of >5 mm Hg after SLT. The majority of anterior uveitis settle within 3 to 5 days of treatment.12

Lee et al12 investigated 111 eyes with OAG that had SLT treatment. The endothelial cell count (specular microscopy), central corneal thickness (CCT; videokeratography), and spherical equivalent (kerato-refractometer) were measured before and at 1 month after SLT. The mean endothelial cell count apparently decreased by 4.5% (from 2465.0 ± 334.0 cells/mm² to 2355.0 ± 387.0 cells/mm²; P=0.0004) during the first week but increased back to baseline level by 1 month. This was due to the attachment of inflammatory cells on the endothelium or a microscopic cellular oedema separating the endothelial cells from the Descemet’s, impairing the accurate counting of endothelial cells.12 In PACG that received SLT to at least 180° of visible trabecular meshwork, it was reported that at 6 months, the endothelial cell count loss was 4.8% (P=0.001).8 The differences between the permanence of endothelial cell damage between the two studies may be related to the closer proximity of the cornea to the trabecular meshwork in PACG subjects. Patients with PACG should be made aware of this potential damage to the endothelium from laser heat dissipation.

There was a transient 1.1% decrease in CCT at 1 week after SLT (from 549.4 ± 37.6 to 543.9 ± 40.2 µm; P=0.02) likely from temporary thermal contractions of the stromal collagen fibres. There was no change in the spherical equivalent but the mean vision after SLT was interestingly reported to improve from 0.3 logMAR to 0.2 logMAR (P<0.0003), possibly related to subjective variation in visual testing and not directly related to the laser itself.12

Intra-ocular spikes, uveitis, and corneal changes are only some of the potential side-effects of SLT that have been reported in studies involving the Hong Kong Chinese population. Although a full review of the side-effects of SLT is beyond the scope and focus of this review, a comprehensive summary of side-effects with incidences can be found in a recent meta-analysis by Wong et al.24

At present, there is no effective means to measure the continuous 24-hour IOP profile in the gold-standard measurement of mm Hg but IOP fluctuation has been demonstrated as a potential risk factor in glaucoma progression.25 26 The SENSIMED Triggerfish (Sensimed AG, Lausanne, Switzerland) is a wireless silicon contact lens sensor (CLS) that can measure the biodimensional changes at the corneoscleral junction, collecting more than 300 data points every 5 minutes. The IOP is measured in output units of millivolts equivalent, thus, the plotted data can only represent the degree of fluctuation from the individual’s baseline pressure.27

Lee et al27 utilised the CLS to measure IOP-related pattern changes in a group of Chinese NTG subjects treated with SLT. The CLS was worn for 24 hours before and for 1 month after the laser procedure while keeping the same number of antiglaucoma medications. A cosine function was fitted to the mean CLS pattern and global variability was analysed in subjects that had success and non-successful to laser, where success was defined as IOP reduction of ≥20%. Local variability from the mean curve was also measured at the diurnal, nocturnal, and 24-hour periods.27 In 44% of subjects who had successful SLT, the global amplitude was reduced by 24.6% but in the non-successful group, subjects experienced a 19.2% increase in their global amplitude, which was primarily driven by a 34.1% greater diurnal local variability.27

Whether or not SLT affects IOP fluctuation remains controversial in the literature for other populations with different authors reporting differences in IOP fluctuation and at different periods of the day.28 29 The majority of subjects in Lee et al’s study29 were prescribed an evening dose of prostaglandin analogues and it has been previously reported that the peak effect of the drug occurs between 8 and 12 hours.30 It seems that in our population, patients who respond to SLT may also benefit from a dampening of their 24-hour IOP-related pattern amplitude although larger-scale studies with 24-hour IOP monitoring in mm Hg would be required to draw more solid conclusions about SLT and IOP fluctuation.

Based on literature related to the Hong Kong Chinese population, SLT is an effective modality for lowering IOP in patients with POAG, OHT, PACG, and NTG. The response or success rate to laser and the amount of IOP reduction varies depending on the type of glaucoma. A higher pre-laser IOP was associated with greater success and delivering a greater total energy seemed to improve the level of IOP reduction. Adverse effects including IOP spikes and uveitis were usually transient although permanent endothelial cell loss may be seen after SLT in patients with anatomically narrow angles. Further research is warranted to assess the repeatability of SLT, long-term sustainability, influence on IOP fluctuation, and role as primary treatment for glaucoma.

1. Lai JS, Chua JK, Tham CC, Lam DS. Five-year follow up of selective laser trabeculoplasty in Chinese eyes. Clin Experiment Ophthalmol 2004;32:368-72. Crossref

2. Traverso CE, Spaeth GL, Starita RJ, Fellman RL, Greenidge KC, Poryzees E. Factors affecting the results of argon laser trabeculoplasty in open-angle glaucoma. Ophthalmic Surg 1986;17:554-9.

3. Schwartz AL, Love DC, Schwartz MA. Long-term follow-up of argon laser trabeculoplasty for uncontrolled open-angle glaucoma. Arch Ophthalmol 1985;103:1482-4. Crossref

4. Wise JB. Ten year results of laser trabeculoplasty. Does the laser avoid glaucoma surgery or merely defer it? Eye (Lond) 1987;1:45-50. Crossref

5. Lee JW, Chan CW, Wong MO, Chan J, Li Q, Lai JS. A randomized control trial to evaluate the effect of adjuvant selective laser trabeculoplasty versus medication alone in primary open-angle glaucoma: preliminary results. Clin Ophthalmol 2014;8:1987-92. Crossref

6. Lee JW, Gangwani RA, Chan JC, Lai JS. Prospective study on the efficacy of treating normal tension glaucoma with a single session of selective laser trabeculoplasty. J Glaucoma 2015;24:77-80. Crossref

7. Ho CL, Lai JS, Aquino MV, et al. Selective laser trabeculoplasty for primary angle closure with persistently elevated intraocular pressure after iridotomy. J Glaucoma 2009;18:563-6. Crossref

8. Narayanaswamy A, Leung CK, Istiantoro DV, et al. Efficacy of selective laser trabeculoplasty in primary angle-closure glaucoma: a randomized clinical trial. JAMA Ophthalmol 2015;133:206-12. Crossref

9. Stein JD, Challa P. Mechanisms of action and efficacy of argon laser trabeculoplasty and selective laser trabeculoplasty. Curr Opin Ophthalmol 2007;18:140-5. Crossref

10. Samples JR, Alexander JP, Acott TS. Regulation of the levels of human trabecular matrix metalloproteinases and inhibitor by interleukin-1 and dexamethasone. Invest Ophthalmol Vis Sci 1993;34:3386-95.

11. Bradley JM, Anderssohn AM, Colvis CM, et al. Mediation of laser trabeculoplasty-induced matrix metalloproteinase expression by IL-1beta and TNFalpha. Invest Ophthalmol Vis Sci 2000;41:422-30.

12. Lee JW, Chan JC, Chang RT, et al. Corneal changes after a single session of selective laser trabeculoplasty for open-angle glaucoma. Eye (Lond) 2014;28:47-52. Crossref

13. Lee JW, Wong MO, Liu CC, Lai JS. Optimal selective laser trabeculoplasty energy for maximal intraocular pressure reduction in open-angle glaucoma. J Glaucoma 2015;24:e128-31. Crossref

14. Iwase A, Suzuki Y, Araie M, et al. The prevalence of primary open-angle glaucoma in Japanese: the Tajimi Study. Ophthalmology 2004;111:1641-8. Crossref

15. Kim CS, Seong GJ, Lee NH, Song KC; Namil Study Group, Korean Glaucoma Society. Prevalence of primary open-angle glaucoma in central South Korea the Namil study. Ophthalmology 2011;118:1024-30. Crossref

16. Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures. Collaborative Normal-Tension Glaucoma Study Group. Am J Ophthalmol 1998;126:487-97. Crossref

17. Lee JW, Ho WL, Chan JC, Lai JS. Efficacy of selective laser trabeculoplasty for normal tension glaucoma: 1 year results. BMC Ophthalmol 2015;15:1. Crossref

18. Schwartz AL, Whitten ME, Bleiman B, Martin D. Argon laser trabecular surgery in uncontrolled phakic open angle glaucoma. Ophthalmology 1981;88:203-12. Crossref

19. Song J, Lee PP, Epstein DL, et al. High failure rate associated with 180 degrees selective laser trabeculoplasty. J Glaucoma 2005;14:400-8. Crossref

20. Lee JW, Shum JJ, Chan JC, Lai JS. Two-year clinical results after selective laser trabeculoplasty for normal tension glaucoma. Medicine (Baltimore) 2015;94:e984. Crossref

21. Lee JW, Liu CC, Chan J, Wong RL, Wong IY, Lai JS. Predictors of success in selective laser trabeculoplasty for primary open angle glaucoma in Chinese. Clin Ophthalmol 2014;8:1787-91. Crossref

22. Lee JW, Liu CC, Chan JC, Lai JS. Predictors of success in selective laser trabeculoplasty for Chinese open-angle glaucoma. J Glaucoma 2014;23:321-5. Crossref

23. Lee JW, Liu CC, Chan JC, Lai JS. Predictors of success in selective laser trabeculoplasty for normal tension glaucoma. Medicine (Baltimore) 2014;93:e236. Crossref

24. Wong MO, Lee JW, Choy BN, Chan JC, Lai JS. Systematic review and meta-analysis on the efficacy of selective laser trabeculoplasty in open-angle glaucoma. Surv Ophthalmol 2015;60:36-50. Crossref

25. Asrani S, Zeimer R, Wilensky J, Gieser D, Vitale S, Lindenmuth K. Large diurnal fluctuations in intraocular pressure are an independent risk factor in patients with glaucoma. J Glaucoma 2000;9:134-42. Crossref

26. Collaer N, Zeyen T, Caprioli J. Sequential office pressure measurements in the management of glaucoma. J Glaucoma 2005;14:196-200. Crossref

27. Lee JW, Fu L, Chan JC, Lai JS. Twenty-four-hour intraocular pressure related changes following adjuvant selective laser trabeculoplasty for normal tension glaucoma. Medicine (Baltimore) 2014;93:e238. Crossref

28. Kóthy P, Tóth M, Holló G. Influence of selective laser trabeculoplasty on 24-hour diurnal intraocular pressure fluctuation in primary open-angle glaucoma: a pilot study. Ophthalmic Surg Lasers Imaging 2010;41:342-7. Crossref

29. Lee AC, Mosaed S, Weinreb RN, Kripke DF, Liu JH. Effect of laser trabeculoplasty on nocturnal intraocular pressure in medically treated glaucoma patients. Ophthalmology 2007;114:666-70. Crossref

30. Russo A, Riva I, Pizzolante T, Noto F, Quaranta L. Latanoprost ophthalmic solution in the treatment of open angle glaucoma or raised intraocular pressure: a review. Clin Ophthalmol 2008;2:897-905.