A previously healthy 48-year-old woman developed acute respiratory distress during operative hysteroscopy for submucosal myomectomy. The preoperative chest X-ray was normal (Fig a). During surgery, facial oedema and acute dyspnoea developed following intrauterine irrigation with 19 100 mL normal saline. Analysis of arterial blood showed mixed metabolic acidosis and respiratory acidosis with pH 6.989, PaCO2 57.8 mm Hg, and a bicarbonate concentration of 14.0 mmol/L (Table). Desaturation developed progressively and her SpO₂ was 85% with O₂ mask at a flow rate of 10 L/min. The PaO₂/FiO₂ ratio was 121.5 mm Hg.

On arrival in the intensive care unit, the patient had a blood pressure of 160/90 mm Hg, heart rate 72 bpm, respiratory rate 30 breaths per minute, and SpO₂ 84%. Chest auscultation revealed bilateral crackles and point-of-care ultrasonography showed a bilateral diffuse B-line pattern consistent with acute pulmonary oedema. Cardiac ultrasound revealed preserved left ventricular systolic function (75.01%) [Fig b]. There was neither regional wall motion abnormality nor aortic stenosis. Non-invasive positive pressure ventilation with bi-level positive airway pressure was commenced to assist ventilation. Subsequent chest X-ray showed acute pulmonary oedema (Fig c) and intravenous diuretic (furosemide, 40 mg) was administered; nevertheless her PaO₂/FiO₂ ratio dropped to 80.5 mm Hg (Table). The overall presentation was compatible with gynaecological transurethral resection of the prostate (TURP) syndrome–induced acute pulmonary oedema.

Intravenous high-dose 1 mg nitro-glycerine (NTG) was administered followed by 1 mg NTG bolus 5 minutes later and continuous infusion at a rate of 1 mg/h. Transient hypotension was noted after the high-dose NTG bolus. The patient’s respiratory distress and oxygenation improved gradually and arterial blood gas analysis revealed improvement with pH 7.304, PaCO₂ 46.9 mm Hg, PaO₂ 64.4 mm Hg, and lactate 38 mg/dL (Table). The next day, chest X-ray revealed resolution of bilateral pulmonary congestion (Fig d). The patient was maintained in a negative fluid balance and fitted with a nasal cannula that was well tolerated. She was transferred to the general ward the next day and discharged from hospital on the third day after surgery.

Hysteroscopic myomectomy is a surgical treatment for submucosal myoma. During hysteroscopy, continuous fluid irrigation is required to distend the uterine cavity for optimal visualisation of the operative field. However, when intrauterine pressure exceeds mean arterial pressure, rapid intravasation via the venous sinuses may occur and absorption of excessive irrigation fluid may cause hypervolaemia and noncardiogenic pulmonary oedema. This phenomenon is termed operative hysteroscopy intravascular absorption syndrome with pathogenesis similar to that of TURP syndrome; as such, operative hysteroscopy intravascular absorption syndrome is also known as gynaecological TURP syndrome.1

Irrigation with electrolyte-free fluid in monopolar operative hysteroscopy may result in dilutional hyponatraemia and hypo-osmolality. Treatment of severe hyponatraemia-related neurological symptoms such as convulsion or cerebral oedema requires an infusion of 3% hypertonic saline solution. Use of a glycine-based solution can result in hyperammonaemia since ammonia is a by-product of glycine once absorbed and metabolised in the liver. In bipolar operative hysteroscopy, 0.9% normal saline is the commonly used irrigant, so there is no significant procedure-related hyponatraemia. Nonetheless in our patient, severe metabolic acidosis developed due to rapid absorption of chloride causing smaller plasma strong ion difference, leading to an increase in hydrogen ion, and therefore acidosis. There are no definitive diagnostic criteria for gynaecological TURP syndrome; the signs are varied and non-specific. Clinical symptoms following a hysteroscopic procedure in patients should be evaluated carefully to exclude or diagnose gynaecological TURP syndrome.

An intravenous NTG infusion is one of the treatments for acute pulmonary oedema. Nitro-glycerine produces nitric oxide that increases the formation of cyclic guanosine 3’,-5’-monophosphate.2 The increased intracellular cyclic guanosine 3’,-5’-monophosphate inhibits the influx of calcium into cells thereby decreasing intracellular calcium levels and causing vascular smooth muscle relaxation. This vasodilator effect on veins causes a decreased venous return, helps reduce cardiac preload and reduces pulmonary capillary hydrostatic pressure. The vasodilator effect on arteries in turn reduces systemic vascular resistance, helps reduce afterload and increases cardiac output. The main effect of NTG is on fluid redistribution of blood volume away from the lungs rather than on fluid removal per se.

To treat acute heart failure, the initial dose of NTG is 10 to 20 μg/min titrated to a total dose up to 200 μg/min until the desired haemodynamic effect is obtained.3 The main side-effects of NTG are hypotension and headache. The major concern of NTG therapy is resistance and tolerance under continuous infusion. In a previous study, intravenous bolus of high-dose NTG (2 mg) with repeated administration every 3 minutes, up to a total of 10 doses in treatment of severe decompensated heart failure, was associated with lower frequency of endotracheal intubation and intensive care unit admissions as well as few adverse events.4 The contra-indications to high-dose NTG bolus are severe aortic stenosis, hypertrophic obstructive cardiomyopathy, hypotension, and concurrent use of phosphodiesterase type 5 inhibitors.

Other strategies to treat acute pulmonary oedema include supplemental oxygen and/or ventilatory support, morphine, diuretics (eg, loop diuretic or thiazide), fluid restriction, inotropic agents in cases of co-morbid hypotension or hypoperfusion, and treatment of any underlying aetiologies.5 Non-invasive positive pressure ventilation improves pulmonary compliance and reduces atelectasis. This can reduce the work of breathing and decrease the need for endotracheal intubation.

The effect of NTG is rapid with an immediate peak effect and short half-life of 3 to 5 minutes. It offers greater efficacy, safety, and a faster onset of action than diuretics or morphine administered to cause rapid improvement of pulmonary congestion. In our case, the patient did not experience significant improvement after diuretics so an intravenous bolus of high-dose NTG was administered along with bi-level positive airway pressure support. As a result, while awaiting the fluid removal effect of diuretic therapy, high-dose NTG helped improve the respiratory distress and avoid an immediate need for intubation.

Our case findings suggest that high-dose NTG can be used to treat gynaecological TURP syndrome–induced acute pulmonary oedema and obviate the need for endotracheal intubation. As far as we know, this is the first case report of the application of high-dose NTG to treat gynaecological TURP syndrome. The success in this case may offer new insight for recommended treatment and further research.

1. Jackson S, Lampe G. Operative hysteroscopy intravascular absorption syndrome. West J Med 1995;162:53-4.

2. Ignarro LJ, Lippton H, Edwards JC, et al. Mechanism of vascular smooth muscle relaxation by organic nitrates, nitrites, nitroprusside and nitric oxide: evidence for the involvement of S-nitrosothiols as active intermediates. J Pharmacol Exp Ther 1981;218:739-49.

3. Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur J Heart Fail 2016;18:891-975. Crossref

4. Levy P, Compton S, Welch R, et al. Treatment of severe decompensated heart failure with high-dose intravenous nitroglycerin: a feasibility and outcome analysis. Ann Emerg Med 2007;50:144-52. Crossref

5. Wilson SS, Kwiatkowski GM, Millis SR, Purakal JD, Mahajan AP, Levy PD. Use of nitroglycerin by bolus prevents intensive care unit admission in patients with acute hypertensive heart failure. Am J Emerg Med 2017;35:126-31. Crossref

Discordant thyroid function test results are a commonly encountered diagnostic challenge. When the inverse log-linear relationship between thyroid stimulating hormone (TSH) and free thyroxine (fT4) is absent, a wide range of differential diagnoses should be considered. This requires the investigatory efforts of endocrinologists and chemical pathologists. We report the diagnostic journey of two Chinese patients with discordant thyroid function test due to familial dysalbuminaemic hyperthyroxinaemia (FDH) [OMIM #615999].

Patient 1 was a woman with no known family history of thyroid disorder. In 2003, she presented at age 49 years with palpitations and weight loss and was diagnosed with thyrotoxicosis (laboratory test results not available). She was prescribed an antithyroid drug for 18 months and her disease was in apparent remission. She presented again in December 2009, at age 55 years, with weight loss and anxiety. Total thyroxine tested in the private sector was elevated at 186.4 nmol/L (reference interval 58-155 nmol/L; platform not specified). She was commenced on carbimazole in view of her “relapse” and referred to endocrinology clinic for further management in March 2010. Her first paired TSH and fT4 (Roche Elecsys) results in April 2011 during carbimazole therapy revealed a discordant pattern with a normal TSH of 1.90 mIU/L and an elevated fT4 at 23.5 pmol/L (Table 1). Due to the apparent “persistent thyrotoxicosis”, she underwent radioactive iodine (RAI) thyroid ablation in May 2011, and carbimazole was stopped after her fT4 was “normalised”. Monitoring with fT4 only, her disease appeared to be in remission. Nonetheless her next paired thyroid function test in May 2013 showed elevated TSH 41.8 mIU/L and low fT4 11.9 pmol/L, consistent with frank post-RAI hypothyroidism. As a result, she was prescribed thyroxine replacement and her TSH was normalised (3.28 mIU/L) by June 2014. She was subsequently followed up in the general outpatient clinic. She was referred again to the endocrine clinic in June 2016 because both TSH and fT4 were elevated despite good compliance with thyroxine treatment. Withholding thyroxine replacement led to elevation of TSH (19.8 mIU/L) while fT4 (15.0 pmol/L) and free triiodothyronine (3.71 pmol/L) were within their respective reference intervals. Chemical pathologists were consulted for discordant thyroid function test results. This specimen was also tested using the Abbott Architect platform and showed similar thyroid function test results. Considering her TSH levels were never suppressed, TSH-dependent thyrotoxicosis was suspected although her serum alpha-subunit level was normal. She was then referred for genetic testing.

Patient 2 was a 34-year-old man who, in February 2004, presented with unintentional weight loss. His thyroid function test revealed normal TSH but elevated fT4 (laboratory test results not available). At follow-up examination, the patient’s TSH was re-checked (1.58 mIU/L; Roche Elecsys) but not his fT4, and test results for antithyroglobulin and antimicrosomal antibodies were negative. In view of such normal results, no further follow-up was arranged until the patient had another thyroid function test on Beckman Coulter Access. This was tested during a preoperative assessment in February 2017 and revealed an elevated fT4 (36.3 pmol/L) with a non-suppressed TSH (1.29 mIU/L). He was referred to the endocrinology clinic for further assessment. The patient reported a family history of thyroid disease: his mother had thyrotoxicosis treated with RAI, a maternal uncle had an unspecified thyroid illness, and an elder brother demonstrated a similar thyroid function test pattern. After discussion with chemical pathologists, subsequent samples were tested on other platforms and showed variable elevation of fT4 (Table 2). Among the four platforms, the degree of elevation in fT4 was greatest on Beckman Coulter Access (209%-224% of upper reference limit), followed by Roche Elecsys (155%). Siemens ADVIA Centaur (101%), and Abbott Architect (105%). The results from all four platforms were above the upper reference limit. Free triiodothyronine was also elevated on Beckman Coulter Access (108%), Abbott Architect (107%) and Siemens ADVIA Centaur (103%) platforms, although to a lesser degree. In view of the non-suppressed TSH level, the possibility of a TSH-secreting tumour was considered. Magnetic resonance imaging showed a 2-mm hypo-enhancing anterior pituitary lesion. Biochemical investigations showed normal serum alpha-subunit level and thyrotropin releasing hormone stimulation test. The patient was referred for genetic testing.

Genetic testing for the THRB gene (OMIM *190160) was performed in both patients, in view of possible resistance to thyroid hormone syndrome (OMIM #188570). No pathogenic variant was detected by polymerase chain reaction and Sanger sequencing. Further testing in both patients for FDH targeting exon 7 of the ALB gene (OMIM *103600; Refseq NG_009291.1/NM_000477.6/NP_000468.1) showed heterozygous c.725G>A p.(Arg242His), a reported pathogenic variant in Chinese.1

Familial dysalbuminaemic hyperthyroxinaemia is an autosomal dominant condition caused by variants of albumin, the gene product of ALB.2 The prevalence of FDH has been estimated to be 0.01% in Caucasian populations but much higher (1.0%-1.8%) in Hispanic populations.3 The prevalence is uncertain in East Asian populations. The variant found in our patients, c.725G>A, has an allele frequency of 0.005437% (1 in 18 392) in East Asian populations, according to the Genome Aggregation Database. However, FDH is less commonly observed than expected in Hong Kong. The under-recognition of FDH may be due to its asymptomatic nature, and the use of TSH without fT4 for screening of thyroid dysfunction. Our patients were identified only because of their apparent thyrotoxic symptoms and testing of fT4 on an affected platform. Although no treatment is required for FDH, it remains an important entity to recognise due to the potentially devastating complications of inappropriate treatment such as iatrogenic hypothyroidism as in patient 1, and adverse effects of antithyroid medications such as agranulocytosis. Family cascade screening and counselling should be considered to prevent undesirable misinterpretation and unnecessary treatment of other affected family members. It is strongly discouraged to treat either TSH or fT4 blindly, as is sometimes observed in clinical practice. Once these patients are treated with antithyroid drugs, often with only fT4 used for monitoring, there is an added risk to the development of frank hypothyroidism as the actual fT4 level is lower than the assayed value.

Here we discuss the approach to an elevated fT4 with a non-suppressed TSH in adults, which often requires the concerted efforts of a chemical pathologist and endocrinologist (Table 3). For a more comprehensive review of the general approach to discordant thyroid function test, the reader is advised to consult the excellent review by Koulouri et al.3 Possible differential diagnoses of an elevated fT4 with a non-suppressed TSH in an adult patient include assay interference (including FDH), drugs (such as thyroxine replacement, amiodarone, heparin), non-thyroidal illness (including acute psychiatric disorders), TSH-secreting pituitary adenoma, and resistance to thyroid hormone.3

First, clinical correlation of thyroid status with other clinical signs is essential in the interpretation of such thyroid function pattern. It would be sensible to exclude conditions that can be examined by a clinical or drug history, such as the effects of drug therapy, acute psychiatric episodes or non-thyroidal illness. A family history of thyroid disorder should be explored, because it may suggest an inherited condition, such as in patient 2.

Second, any previous thyroid function results should be reviewed. In patients with thyroid function test results incompatible with clinical features, at least one paired TSH and fT4 and/or free triiodothyronine should be performed to better detect discordant results. Patients who have previously shown a normal thyroid function pattern are less likely to have inherited conditions such as resistance to thyroid hormone or FDH. For FDH, it is important to note the assay platform used (hinted by the reference intervals provided), as a change of assay from one that is less affected by FDH to one that is prone to interference in FDH may give the clinician a false sense of an acquired condition. Moreover, the TSH level in patients with FDH should be normal unless due to thyroid-related treatment, or genuine pituitary or thyroid disease. It is often useful to plot the TSH and fT4 values and try to observe the reciprocal relationship between them. In FDH patients, such a relationship should be largely preserved even after antithyroid treatment, despite an upward shift in fT4 (Fig). This suggests the presence of a functional negative feedback system with the hypothalamus-pituitary-thyroid axis and is less likely to be a case of interference. Thyroxine replacement may further complicate the thyroid function test pattern if thyroxine is taken shortly before blood taking, although the usual thyroid function test requested for patients receiving thyroxine replacement is only TSH.

Third, further investigations for possible assay interference should be discussed with chemical pathologists. In most clinical laboratories, thyroid function tests are performed using automated immunoassays that are prone to interference, for example by macro-TSH, biotin, anti-streptavidin antibodies, anti-ruthenium antibodies, thyroid hormone autoantibodies or heterophilic antibodies. Further testing such as assay comparison, serial dilution, antibody blocking agents, and polyethylene glycol precipitation may be considered to provide useful clues for possible interference.4 Often either the TSH or fT4 assay is affected. Nonetheless, in some cases such as biotin interference, both TSH and fT4 can be affected if they both use the biotin-streptavidin-based separation method in the assay. In patients with FDH, due to an increased binding affinity of thyroxine to albumin, total thyroxine is elevated, and to a lesser extent total triiodothyronine, although fT4 measured using equilibrium dialysis, which only minimally disrupts the equilibrium between the free and the bound portion of thyroxine, should be normal.4 Nevertheless, the presence of this albumin variant has been shown to lead to overestimation of fT4 levels in Roche Elecsys and Siemens Immulite (competitive one-step assays), whereas Abbott Architect (two-step assay) is less affected, because the thyroxine analogue does not come into contact with the variant albumin.5 Beckman Coulter Access, a two-step assay, is an exception as it also shows significant overestimation of fT4. The practical implication is that when FDH is suspected, one should examine the type of fT4 assay used, and repeat testing on a less affected platform.

Apart from the usual investigations for TSH-secreting pituitary adenoma such as magnetic resonance imaging of the pituitary and serum alpha subunit to TSH molar ratio, or genetic testing for resistance to thyroid hormone, further biochemical testing of the thyroid status of the patient should be considered. Tissue markers of thyroid hormone hyperfunction such as increased sex hormone-binding globulin, ferritin, pro-collagen-1-N-terminal peptide, and decreased cholesterol may provide additional evidence of a thyrotoxic state.6 The absence of these may suggest resistance to thyroid hormone, or assay interference. Further investigations such as serum thyroglobulin, thyroid antibodies (including anti-TSH receptor antibodies), thyroid scan, and ultrasonography may provide additional clues to the thyroid status in difficult cases. However, their use should be discussed with endocrinologists or relevant specialists to ensure optimal test utilisation and interpretation.

In summary, although both our patients ultimately received the correct diagnosis of FDH after collaboration between endocrinologists and chemical pathologists, the complications associated with misinterpretation and inappropriate treatment illustrate the importance of a good understanding of the analytical pitfalls of thyroid function test assays and proper follow-up investigations and management of discordant thyroid function test. Clinicians should be aware of the approach to discordant thyroid function, especially those who treat thyroid conditions.

Concept or design: All authors.
Acquisition of data: All authors.
Analysis or interpretation of data: NKC Lau, TKC Tsui, JSS Kwok.
Drafting of the article: NKC Lau, TKC Tsui, JSS Kwok.
Critical revision for important intellectual content: All authors.

All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

We would like to thank all laboratory and clinical staff who assisted in the cross-platform testing and management of the patients.

This case report received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Informed consent was obtained from all patients.

1. Tiu SC, Choi KL, Shek CC, Lau TC. A Chinese family with familial dysalbuminaemic hyperthyroxinaemia. Hong Kong Med J 2003;9:464-7.

2. Kragh-Hansen U, Galliano M, Minchiotti L. Clinical, genetic, and protein structural aspects of familial dysalbuminemic hyperthyroxinemia and hypertriiodothyroninemia. Front Endocrinol (Lausanne) 2017;8:297. Crossref

3. Koulouri O, Moran C, Halsall D, Chatterjee K, Gurnell M. Pitfalls in the measurement and interpretation of thyroid function tests. Best Pract Res Clin Endocrinol Metab 2013;27:745-62. Crossref

4. Favresse J, Burlacu MC, Maiter D, Gruson D. Interferences with thyroid function immunoassays: clinical implications and detection algorithm. Endocr Rev 2018;39:830-50. Crossref

5. Cartwright D, O’Shea P, Rajanayagam O, et al. Familial dysalbuminemic hyperthyroxinemia: a persistent diagnostic challenge. Clin Chem 2009;55:1044-6. Crossref

6. Singh BK, Yen PM. A clinician’s guide to understanding resistance to thyroid hormone due to receptor mutations in the TRα and TRβ isoforms. Clin Diabetes Endocrinol 2017;3:8. Crossref

A 57-year-old man underwent cadaveric renal transplantation in January 2018 and was prescribed mycophenolate mofetil, tacrolimus, and prednisolone as post-transplant immunosuppressive therapy. He developed multiple cervical lymphadenopathies at 6 months after transplantation. Fine needle aspiration cytology of the left submandibular lymph node, performed in the private sector, revealed only a hypocellular lesion. Considering the possibility of a post-transplantation lymphoproliferative disorder, excisional biopsy was arranged in our unit, but the patient defaulted from his appointment.

The patient attended the emergency department 9 months after transplantation complaining of shortness of breath. Physical examination on admission revealed generalised lymphadenopathy. Chest radiograph showed left lower, left middle, and right lower zone infiltrates. Despite the use of empirical piperacillin/tazobactam and withholding of mycophenolate mofetil, his condition deteriorated with worsening type one respiratory failure and increasing bilateral lung infiltrates on serial chest radiographs. Tacrolimus was discontinued. However, serial procalcitonin levels were undetectable. Computed tomography scan of the thorax suggested focal consolidative changes with diffuse cervical, axillary, mediastinal, hilar and abdominal lymphadenopathy (Fig 1). Biopsy of the groin and cervical lymph node showed spindle cell proliferation associated with surrounding vascular channels, red cell extravasation between spindle cells (Fig 2a), and positive human herpesvirus 8 (HHV-8) staining (Fig 2b). This confirmed the diagnosis of Kaposi’s sarcoma (KS). There was no plasmablastic histopathology to suggest the presence of multi-centric Castleman disease. Despite maximum supportive therapy with mechanical ventilation, empirical antimicrobials and antifungal treatment, his condition further deteriorated and he succumbed due to respiratory failure.

Immunosuppressive therapy is known to increase the risk of infection and malignancy. In a study of incidence of malignancy among a cohort of Hong Kong kidney transplant recipients from 1972 to 2011, the most prevalent malignancies were non-Hodgkin’s lymphoma followed by colorectal cancer, lung cancer, kidney cancer, and non-melanoma skin cancer. Only five cases of KS were reported.1 The low incidence of KS may be due to the low prevalence of HHV-8 seroprevalence in Asia.2 There were only 68 reports of KS in Hong Kong between 1983 and 2016 according to the Hong Kong Cancer Registry. In this report, we describe a case of KS in a renal transplant recipient who presented with multiple cervical lymphadenopathies.

In the clinical context of multiple cervical lymphadenopathies in a post-transplant recipient, post-transplant lymphoproliferative disorder, and multi-centric Castleman disease are our initial top differential diagnoses. However, the histology of the lymph node of our patient did not suggest these diagnoses but KS. The clinical presentation of post-transplant KS can be divided into exclusive dermatological lesions or mucocutaneous lesions, with or without visceral involvement. Most patients present with single or multiple pigmented skin lesions with or without lower limb skin lymphoedema.3 4 5 6 7 Non-cutaneous KS is uncommon and was reported to account for only 5.4% of KS in a large AIDSassociated KS cohort.8 The gastrointestinal tract and the lungs are the most common sites for visceral involvement.4 7 Lymph node involvement is commonly associated with diffuse dermatological lesions or visceral involvement.7 9 Concomitant lymphoma is a possible but uncommon diagnosis for multiple cervical lymphadenopathies in a patient with KS.5 Overall, post-transplant KS that presents with multiple cervical lymphadenopathies without skin lesions is rare.

The first case of KS after renal transplantation was reported in 1969. Its prevalence has been reported as 0.4% to 5.3%, depending on the geographical prevalence of HHV-8 seropositivity.10 11 The average time to diagnosis has been reported as 12 to 39 months after transplantation.3 4 10 There is a male preponderance with a male-to-female ratio of 3:1.4 The 5-year survival rate is around 70%, although those with visceral involvement generally carry a poorer prognosis.12

There are no established guidelines for treatment of post-transplant KS so treatment often depends on clinical presentation. Tapering or withdrawal of immunosuppressants is the mainstay of therapy. Intralesional chemotherapy may be used for a single dermatological lesion. Systemic chemotherapy, such as liposomal anthracycline or taxanes, may be considered for widespread disease. Complete remission with immunosuppressant reduction or withdrawal has been reported as 50% to 60% and graft loss as 20% to 30% in the pre-mammalian target of rapamycin (mTOR) inhibitor era.7 10 Treatment with mTOR inhibitor, such as sirolimus, has gained recognition with its anti-angiogenic and anti-neoplastic activity, particularly among patients with an exclusive dermatological presentation. Three previous case series of 25 patients reported a 100% remission rate after switching from a calcineurin inhibitor to sirolimus, while two patients had graft loss due to causes other than rejection.3 5 6 Other case series have reported treatment failure with sirolimus. Visceral involvement and delay in switching from calcineurin inhibitor to mTOR inhibitor after diagnosis of KS may contribute to treatment failure.9 Further study is essential to determine the optimal treatment for post-transplant KS, especially for those with visceral involvement.

Post-transplant KS presenting with multiple cervical lymphadenopathies is rare and may signify an aggressive subtype. Withdrawing immunosuppressive therapy alone failed to salvage our patient. Further study is required to evaluate the potential of switching mycophenolate mofetil and calcineurin inhibitor to mTOR inhibitor to improve the prognosis for this subgroup of patients.

All authors contributed to the concept of the study, acquisition and analysis of the data, drafting of the manuscript, and critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

We thank Dr Matthew KL Tong, Dr Hilda WH Chan, Dr Hon-lok Tang, and Dr Samuel KS Fung for their valuable opinions and supervision on writing the manuscript. We thank Dr Yuen-fun Mak for her expert guidance on pathology review.

Part of the content about this case was presented in the Nephrology interhospital meeting in September 2019.

All authors have disclosed no conflicts of interest.

This case report received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The patient was treated in accordance with the Declaration of Helsinki. The patient provided informed consent for all procedures.

1. Cheung CY, Lam MF, Chu KH, et al. Malignancies after kidney transplantation: Hong Kong renal registry. Am J Transplant 2012;12:3039-46. Crossref

2. Zhang T, Shao X, Chen Y, et al. Human herpesvirus 8 seroprevalence, China. Emerg Infect Dis 2012;18:150-2. Crossref

3. Stallone G, Schena A, Infante B, et al. Sirolimus for Kaposi’s sarcoma in renal-transplant recipients. N Engl J Med 2005;352:1317-23. Crossref

4. Penn I. Sarcomas in organ allograft recipients. Transplantation 1995;60:1485-91. Crossref

5. Yaich S, Charfeddine K, Zaghdane S, et al. Sirolimus for the treatment of Kaposi Sarcoma after renal transplantation: a series of 10 cases. Transplant Proc 2012;44:2824-6. Crossref

6. Gutiérrez-Dalmau A, Sánchez-Fructuoso A, Sanz-Guajardo A, et al. Efficacy of conversion to sirolimus in posttransplantation Kaposi’s sarcoma. Transplant Proc 2005;37:3836-8. Crossref

7. Barete S, Calvez V, Mouquet C, et al. Clinical features and contribution of virological findings to the management of Kaposi sarcoma in organ-allograft recipients. Arch Dermatol 2000;136:1452-8. Crossref

8. Stebbing J, Mazhar D, Lewis R, et al. The presentation and survival of patients with non-cutaneous AIDS-associated Kaposi’s sarcoma. Ann Oncol 2006;17:503-6. Crossref

9. Lebbé C, Euvrard S, Barrou B, et al. Sirolimus conversion for patients with posttransplant Kaposi’s sarcoma. Am J Transplant 2006;6:2164-8. Crossref

10. Montagnino G, Bencini PL, Tarantino A, Caputo R, Ponticelli C. Clinical features and course of Kaposi’s sarcoma in kidney transplant patients: report of 13 cases. Am J Nephrol 1994;14:121-6. Crossref

11. Qunibi W, Akhtar M, Sheth K, et al. Kaposi’s sarcoma: the most common tumor after renal transplantation in Saudi Arabia. Am J Med 1988;84:225-32. Crossref

12. Woodle ES, Hanaway M, Buell J, et al. Kaposi sarcoma: an analysis of the US and international experiences from the Israel Penn International Transplant Tumor Registry. Transplant Proc 2001;33:3660-1. Crossref

A 69-year-old man with a history of diabetes mellitus and benign prostatic hypertrophy was admitted to our medical unit with a 1-week history of fever, runny nose, and shortness of breath. Chest X-ray revealed bilateral diffuse ground glass opacities in both lung fields. The patient had type 1 respiratory failure and required non-invasive positive pressure ventilation to maintain oxygenation. The patient was given ceftriaxone 1 g every 12 hours, and doxycycline 100 mg and oseltamivir 75 mg twice daily. He was transferred to the intensive care unit and intubated due to severe acute respiratory distress syndrome (ARDS). The patient’s nasopharyngeal swab was tested by polymerase chain reaction and the result was positive for influenza A virus H1 RNA. His clinical course was further complicated by septic acute kidney injury that necessitated continuous venovenous haemofiltration. His lungs gradually improved with supportive measures including prone ventilation and muscle paralysis for 48 hours. The fractional oxygenation requirement improved from 1.0 to 0.4 between day 1 and day 8 of admission and chest X-ray showed improving aeration of both lungs.

Unfortunately, his ventilatory requirement deteriorated again from day 9 onwards, with chest X-ray showing new right upper zone and left middle zone infiltrates (Fig 1). Repeat microbiological investigation results were negative except for Candida albicans and Candida species in the endotracheal culture. Antibiotics were upgraded to meropenem 500 mg every 8 hours. High-resolution computed tomography of the thorax on day 14 showed patchy consolidation with air bronchogram and ground glass opacities in both lungs, more prominent in the right upper lobe, right lower lobe, and left lower lobe, and mild bilateral pleural effusions.

Bronchoscopy with transbronchial biopsy of the apical segment of the left lower lobe was performed uneventfully. Histological examination of the biopsy specimen revealed alveolar lining partially composed of cuboidal cells with foamy cytoplasm, a variable degree of nuclear enlargement and pleomorphism, and prominent nucleoli. The histology could be suggestive of adenocarcinoma with appropriate clinical settings. In view of the unusual clinical course, it was decided to repeat transbronchial biopsies at different sites, namely at the left lower lobe apical segment again on day 18 and the right upper lobe posterior segment on day 19. Both histological specimens revealed a pattern compatible with organising pneumonia (OP) [Fig 2]. Intravenous hydrocortisone 100 mg every 8 hours and subsequently enteral prednisolone 40 mg daily were prescribed, and surgical tracheostomy performed in view of the prolonged mechanical ventilation.

The patient showed dramatic improvement with steroid treatment. He was weaned off mechanical ventilatory support and renal replacement therapy and discharged from the intensive care unit on day 22. He underwent rehabilitation in a general ward and was weaned off tracheostomy and supplementary oxygen. Serial chest X-rays showed resolution of both lung consolidations with minimal residual fibrotic scarring.

In general, H1 influenza infection causes mild upper respiratory tract symptoms. A minority of patients present with ARDS secondary to viral pneumonia with or without bacterial co-infection. Apart from neuraminidase inhibitors such as oseltamivir, management is mostly supportive. This includes protective lung ventilation and a conservative fluid management strategy. In moderate to severe ARDS with partial pressure of oxygen/fraction of inspired oxygen ratio <150 mm Hg, 48 hours of muscle paralysis, prone ventilation, and extracorporeal membrane oxygenation may be required.

The definition of ARDS includes acute onset within 1 week, bilateral radiological opacities not explained by pleural effusion or lung collapse, respiratory failure not explained by heart failure, or fluid overload with partial pressure of oxygen/fraction of inspired oxygen ratio <300 mm Hg under a positive end-expiratory pressure of at least 5 cm H2O. Typically, patients with ARDS exhibit diffuse alveolar damage in histological specimens, divided into an acute exudative phase shortly after the pulmonary insult, followed by an organising phase, with or without a final fibrotic phase. However, a continuum and overlapping features exist, especially late in the first week or if the patient has encountered repeat pulmonary insults. The acute exudative phase is characterised by hyaline membranes that gradually disappear in the subsequent organising phase. The organising phase is characterised by interstitial fibrosis and pronounced type 2 pneumocyte hyperplasia. Cytological atypia may be quite pronounced and can be confused with malignancy, as in our case.1 Subsequently, diffuse alveolar damage will gradually resolve although some may progress to a fibrotic phase with continued interstitial fibrosis and compromises lung function.

Other histological patterns can present in patients with ARDS, including OP and acute fibrinous OP. In acute fibrinous OP, the alveolar spaces are filled with organising fibrin balls instead of hyaline membranes, whereas in OP, patchy accumulation of intra-alveolar organising fibroblastic tissue primarily centred around bronchioles is present.2

Organising pneumonia, acute fibrinous OP, and diffuse alveolar damage are histological manifestations arising from a broad range of pulmonary insults. Temporal and regional heterogeneity of the pulmonary parenchymal alterations may result in diverse or mixed patterns upon lung biopsy sampling. Corticosteroids remain the first-line therapy in OP. Around two thirds of patients with OP respond to treatment with corticosteroids. However, the role of corticosteroids in diffuse alveolar damage remains controversial. The steroid response of acute fibrinous OP is intermediate, between that of diffuse alveolar damage and OP.3

To the best of our knowledge, few cases of post-influenza OP have been reported in the literature.4 5 Most were influenza A infection and patients usually presented with refractory respiratory failure, incomplete recovery, or new radiological infiltrate after initial improvement. Most cases responded well to corticosteroid therapy.

The present case demonstrates that OP can complicate influenza-related ARDS. Physicians should be aware of this possibility: timely confirmation by histological proof, exclusion of superimposed nosocomial pneumonia, and initiation of corticosteroid therapy after balancing the risks and benefits may result in a more favourable outcome in the disease trajectory.

All authors contributed to the concept of study, acquisition and analysis of data, drafting of the manuscript, and had critical revision of the manuscript for important intellectual content. All authors had full access to the data, contributed to the study, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

This case report received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The patient was treated in accordance with the Declaration of Helsinki. The patient verbally agreed publication of this anonymous case report.

1. Butnor KJ. Avoiding underdiagnosis, overdiagnosis, and misdiagnosis of lung carcinoma. Arch Pathol Lab Med 2008;132:1118-32.

2. Beasley MB. The pathologist's approach to acute lung injury. Arch Pathol Lab Med 2010;134:719-27.

3. Bihari S, Bailey M, Bersten AD. Steroids in ARDS: to be or not to be. Intensive Care Med 2016;42:931-3. Crossref

4. Cornejo R, Llanos O, Fernández C, et al. Organizing pneumonia in patients with severe respiratory failure due to novel A (H1N1) influenza. BMJ Case Rep 2010; 2010. pii: bcr0220102708.Crossref

5. Asai N, Yokoi T, Nishiyama N, et al. Secondary organizing pneumonia following viral pneumonia caused by severe influenza B: a case report and literature reviews. BMC Infect Dis 2017;17:572. Crossref

A 59-year-old Chinese man with a history of follicular lymphoma presented in November 2015 with painless swelling in the right upper back. The patient had no history of trauma. Clinical examination revealed a mild non-tender prominence over the right scapula without palpable axillary lymph nodes. Musculoskeletal and neurovascular examinations were normal.

Plain radiograph of the right shoulder showed a well-circumscribed radiolucent lesion in the body of the right scapula with an adjacent irregular osseous protuberance. Computed tomographic scan revealed a well-defined lipomatous subscapular lesion with internal calcific foci adjacent to the body of the scapula with juxtacortical bony excrescence (Fig 1a and b). There was no continuity with the medulla of the parent bone. The provisional diagnosis was lipoma. Because of the history of follicular lymphoma, ¹⁸F-fluorodeoxyglucose (FDG) positron emission tomography–computed tomography (PET-CT) scans had been performed at regular intervals for 3 years; these scans incidentally showed a serial increase in the size of the right scapular lesion from 2.5 cm × 3.8 cm to 3.2 cm × 5.8 cm. No significant FDG uptake was seen (Fig 1c).

Magnetic resonance imaging (MRI) scan performed for further characterisation showed a well-marginated 3.2-cm × 7.6-cm × 6.6-cm soft tissue mass beneath the subscapularis muscle abutting the subscapularis fossa (Fig 2). Signal intensity was predominantly identical in all pulse sequences to that of subcutaneous fat, suggestive of a lipomatous component. The central part of the lesion was heterogeneous in signal intensity, with T1-weighted and T2-weighted low signal structures extending and radiating from the bony scapula, corresponding to the bony protuberance seen on CT scan. Thin internal septa were noted within the lesion. No cortical or medullary continuation with the parent bone was evident. There was no gross invasion of the bony scapula. No muscle atrophy was seen to suggest nerve entrapment.

While the imaging features were compatible with that of a parosteal lipoma, CT-guided biopsy of the lesion was subsequently performed for confirmation. Under local anaesthesia and CT guidance, the right scapular body was penetrated via a posterior approach through the infraspinatus muscle and the parosteal lipomatous tumour in the subscapular fossa was biopsied multiple times.

Pathology examination showed tissue mainly consisting of mature fat cells interspersed by thin fibrous septa. Cells with enlarged hyperchromatic nuclei were not seen. Some bony fragments were seen. Findings were consistent with a parosteal lipoma. Conservative management was decided in view of the deep-seated location of the lesion. Follow-up MRI scan showed no significant interval change.

Although lipoma is a common benign tumour of mature adipose tissue, which can occur in almost any organ, parosteal lipoma is a rare type of lipoma, accounting for 0.3% of all lipomas.1 Parosteal lipoma is contiguous to the periosteum and is most commonly found near the diaphysis or diametaphysis of the long bones, in particular the femur and proximal radius, and less commonly the tibia, humerus, scapula, clavicle, ribs, pelvis, metacarpals, metatarsals, and mandible.1 2 The scapula is considered a rare site of involvement with only a few cases reported in the literature.

First described by Seering in 1836,3 the term "periosteal lipoma" is strictly defined as a lipoma originating from the periosteum. The term "parosteal lipoma" was coined by Power in 18884 to emphasise that the lesion is adjacent to the bone but does not necessarily arise from it since the periosteum contains no fat cells.

Parosteal lipoma patients, aged 40 to 60 years, usually present with a history of a slow-growing, large, painless, non-tender immobile mass that is not fixed to the skin.5 Motor or sensory deficits from nerve compression caused by parosteal lipomas are common compared with other lipomatous lesions and are most frequently associated with forearm lesions affecting the posterior interosseous nerve.6

The imaging features of parosteal lipoma are usually characteristic. In a study of 32 cases of parosteal lipoma by Fleming et al,1 nearly 70% of patients had abnormal underlying bone and 50% had osseous reaction, both seen in our patient. The most frequent osseous reactive changes, as can be demonstrated in radiograph and CT studies, include bony protuberance, bowing of bone, or cortical pressure erosion secondary to the adjacent lipomatous mass.1 These cortical abnormalities do not display medullary or cortical continuity with the underlying bone.

As illustrated in our case, on MRI images, parosteal lipoma is identified as a juxtacortical mass with signal intensity identical to that of subcutaneous fat, regardless of pulse sequences. T1- and T2-weighted hypointense areas represent osseous components. Areas with intermediate signal intensity on T1-weighted images that are high signal intensity on T2-weighted images represent the cartilaginous components of parosteal lipoma.7 Fibrovascular septa may cause a lobulated appearance of the fat component, with low-signal-intensity strands on T1-weighted images that become higher in signal intensity on the long TR images, especially with fat suppression. Adjacent muscle atrophy is identified as increased striations of fat in the affected muscle and is caused by associated nerve entrapment.7

Although there are no reports of malignant transformation, the treatment of choice for parosteal lipoma is complete surgical excision before irreversible muscle atrophy takes place, especially in cases with nerve entrapment. Parosteal lipomas are characteristically encapsulated and strongly adherent to the underlying periosteum, particularly at sites with osseous proliferation. This often requires subperiosteal dissection or segmental resection of bone.

All authors contributed to the concept of the study, acquisition and analysis of the data, drafting of the article, and critical revision for important intellectual content. All authors had full access to the data, approved the final version for publication, and take responsibility for its accuracy and integrity.

All authors have disclosed no conflicts of interest.

We thank Dr LM Kwok from Queen Elizabeth Hospital for her pathology input.

This case report received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

The patient was treated in accordance with the Declaration of Helsinki. The patient provided informed consent for all procedures.