The ugly side of medical tourism

For surgeon Dr Nicholas Moncrieff and his colleagues at Hunter Plastic Surgery, the sight of women disfigured by botched cosmetic operations overseas is a distressingly common one.

Almost every week the clinic, based in Charleston, just outside Newcastle, gets a new patient upset with the results of sub-standard procedures that have left them with collapsed noses, scarred faces, misaligned breasts and nipples or skewed belly buttons.

Among recent cases was a 40-year-old woman who went to Thailand for cosmetic surgery, including the insertion of a silicon implant in her nose to give it more projection.

It subsequently became badly infected and exposed, leaving her with a serious defect (see attached pictures).

Dr Moncrieff had to operate on her in hospital to remove foreign matter, repair the hole and use filler to fill the cavity.

The surgeon said the vast majority of such cases involved patients who had gone to Thailand for cosmetic surgery, lured by cheap prices.

In Thailand it typically costs around $4000 to $6000 to have breast augmentation surgery, compared with $11,000 at Hunter Plastic Surgery.

But Dr Moncrieff said people who chose where to have surgery based simply on price did not fully realise the sort of risks they were taking.

“I think it [plastic surgery] has been a little trivialised because it has become commonplace,” he said. “This is surgery, it is still an operation on your body, and people think it’s like a haircut.”

Often, cosmetic procedures performed overseas are much more radical than clinics like Hunter Plastic Surgery would undertake, such as inserting very large breast implants in women with small frames, leading to complications such as sore backs or infection because sutures are put under enormous strain.

Dr Moncrieff said any surgery, no matter where it was performed, carried with it the risk of complications, but when it was conducted overseas the burden for rectifying any mistakes fell on the Australian health system.

He said patients were often told by their Thai doctors that any complications could be addressed for free in Australia under Medicare, and public hospitals in his area regularly had to treat women who had become badly infected following cosmetic surgery performed overseas.

But he warned that this did not extend to rectifying elective cosmetic problems, such as facial scarring or nipples in the wrong place.

As an example, the woman who had to have her botched nose surgery repaired spent more than $3000 on the repair work – far more than it would have cost to have the procedure performed locally in the first place. She was able to claim back just $750 of the cost through Medicare.

Dr Moncrieff said it could cost up to $19,000 to repair botched breast augmentation procedures, a sum that was prohibitively expensive for many women who had had such work overseas in the first place because of price, and who usually had no private health cover.

He said in this instance, they often had to live with the disfigurement until they could save the money to have it repaired, either in Australia or in the country where the procedure was originally carried out.

Dr Moncrieff said his concern with the cosmetic tourism trade was not driven by self-interest – “we are plenty busy enough without having to fix these problems” – but by the burden it was placing on taxpayers and the health system.

“We think the Australian Government should be asking question about how complications will be managed [under the NIB scheme] once the patient is back in Australia and who will be paying for them, especially those treated for life-threatening illnesses in public hospitals” he said.

Adrian Rollins

Hand transplantation

Unconventional decisions made by the Australian transplant team provided the best overall outcome for their patient

Hand transplantation represents the spearhead of the new field of composite tissue allotransplantation. Since the beginning of time, it has been a medical goal to replace missing tissue with like tissue. The psychological and medical importance of achieving this goal is highlighted by the story of Cosmas and Damian, two 3rd century physician saints who successfully transplanted a leg around the year 1150, long after their deaths.1

Such reports notwithstanding, the ability to transplant composite tissue remained elusive for most of the history of mankind. The modern era of hand transplantation started in 1964, when Roberto Gilbert Elizalde of Ecuador transplanted a hand onto a patient using azathioprine for immunosuppression. The hand survived for 3 weeks.2

The next attempt to transplant a hand was carried out in France on a New Zealander residing in Australia, Clint Hallam, by a team led by Jean-Michel Dubernard in September 1998.3 Earl Owen from Australia participated in the surgery. The transplanted hand survived for 29 months.

The next hand transplant was carried out in January 1999 by the Kleinert Kutz team in the United States.4 The patient, Matthew Scott, is still using this hand over 14 years later, and this surgery represents the first successful long-term hand transplant.

The success of this hand transplant did not signal medical acceptance of hand transplantation. Many were sceptical about the long-term viability of Matthew Scott’s new hand. The spectre of the short survival of the hand in the first French case, along with the belief that skin was too allogeneic to transplant, generated many critics. There were also those who held the position that hand transplantation was unethical, because one was giving immunosuppression that would shorten the patient’s life, for a transplant that did not aim to extend the patient’s life.

In spite of these concerns, pioneers pushed forward. By the end of the first decade of the new millennium, more teams entered this field. Further hand transplants were carried out in France, the US, Belgium, Austria, Spain and China.5,6 International and American societies for composite tissue transplantation were founded. Other teams started transplanting faces, legs, and combinations of face and hands, hands and legs, tracheas and uteruses. The wider field of composite tissue allotransplantation (CTA) had been established.

The first Australian hand transplant is reported in this issue of the Journal.7 In many ways the report challenges conventional wisdom, and demonstrates the value that this new field can bring to an individual.

Hand transplantation was felt to be dangerous because of the complications it would bring to an otherwise healthy recipient. However, the complications of hand transplantation have actually been less than those seen in solid organ transplantation. This is because hand transplant patients are healthy, whereas solid organ transplant patients are often already severely affected by organ failure and they begin immunosuppression treatment in a poorer state of health.

The Australian case challenges this issue. Conventional wisdom states that all hand transplant patients should be healthy, but this patient had a series of concerning medical issues, including splenectomy and diabetes. The Australian team transplanted a patient who was the oldest recipient on record, with significant medical problems. Was this optimal recipient selection? Many experts in the field of CTA would say no. Should older age be an exclusion criterion? In solid organ transplantation, recipients massively outnumber donors, and allocation criteria have to be firm. However, due to the paucity of hand transplant recipients, there is a relative excess of donors, albeit small. This surplus means we do have more leeway to relax some criteria such as age.

To make matters even more controversial, the authors only transplanted one hand. Immunosuppression partially shuts the immune system down for the whole body. Therefore bilateral hand transplants do not need more immunosuppression than single hand transplants. It would have been possible to transplant both hands without any increase in immunosuppression.

By transplanting a patient who had multiple medical problems, and only transplanting one hand in a patient who arguably needed two, this team challenged prevailing beliefs. However, in my opinion, these were appropriate hand transplant decisions, and it was proper not to transplant the recipient’s only functional partial hand.

The most important goal in hand transplantation is to improve quality of life, and this should not be confused with obtaining the best function. This particular patient had poor quality of life, and because he had lost all four of his extremities, he could perform few activities of daily living. His functional appendage that had been fashioned into a partial hand allowed him only primitive pinch.

The authors saw through the fallacy that all hand transplant patients must be healthy, and that all hands need to be transplanted in cases of bilateral loss. They protected the patient in case of early loss of the transplanted hand, by leaving one partially functional appendage. They correctly understood that, even if older and more prone to complications from immunosuppression, the patient’s quality of life was so poor that a single transplanted hand would provide him with a large increase in utility.

The authors should be congratulated for their clarity of thought in transplanting this patient. “Discovery consists of seeing what everyone has seen and thinking what nobody has thought.”8 This transplant has achieved that. It emphasises the importance of utility over function and health. It opens up the debate about where that balance should be. There will be many who will disagree with these authors and with my words. The patient will not be one of them.

Restless legs syndrome and day procedures

To the Editor: This letter encourages timely preoperative recognition of restless legs syndrome (RLS), for patient wellbeing, as well as for practical management of incidental procedures. RLS is a neurological disorder found in many older patients of either sex and in those with a variety of conditions including iron deficiency, pregnancy, uraemia, diabetes and rheumatoid arthritis. It may also affect patients with other neurological disorders such as peripheral neuropathy.1

Affected persons have spontaneous, recurring, unpleasant leg paraesthesia, which they may find difficult to describe. Patients know that these discomforts are about to climax into sudden involuntary movement of the limb or limbs. RLS occurs mainly in the late afternoon and evening, especially when patients are seated or supine. Sleep may be disturbed. The anticipated movement can be overcome by voluntary activation of the limb or limbs, by bearing weight or by pacing.2 RLS is believed to arise from abnormalities in brain neurotransmitters that regulate automatic movements.

Dopamine receptor agonist medications can suppress the attacks but not cure them.1–4 In patients whose drug therapy is already effective, perioperative stress may bring on RLS, as will dopamine antagonists.1,3 Where RLS is suspected or actual, medications that may exacerbate RLS symptoms should not be used, particularly in day surgery.1,3,4 The following medications should be avoided, whether as premedications, intra-procedural drugs or sedatives:

antiemetic drugs: metoclopramide or prochlorperazine (a phenothiazine)
antipsychotic drugs: droperidol
some anticonvulsants
those antidepressants that increase serotonin levels
lithium
opioid antagonists: naloxone
antihistamines: diphenhydramine
beta-blockers

Surgeons and others who perform procedures (including ophthalmologists, dentists and endoscopists) should ask screening questions before procedures and check for the occurrence of RLS and related sleep disorder. Informed consent should be obtained and pre- and perioperative therapy established. An anaesthetist should be in attendance for the procedure when treating a patient with RLS. All assistants and nursing staff should be informed about this patient’s RLS, and warned that pinning down a patient’s jerking legs may cause reciprocal jerks of the upper torso.

Drugs that relieve RLS should be made available in the procedural theatre or dental practice, eg, intravenous physostigmine 1–2 mg, opioids, oral pramipexole and ropinirole.3–5

(Coincidentally, from February 2013, RLS has been renamed Willis–Ekbom Disease, after the Restless Legs Syndrome Foundation adopted this new title.)

Initial outcomes of using allografts from donation after cardiac death donors for liver transplantation in New South Wales

Despite liver transplantation being an established treatment modality in Australia, there continues to be a significant disparity between donor liver availability and demand.1 One way to reduce this gap is to maximise the use of extended-criteria deceased donor livers, with donation after cardiac death (DCD) being one such option.2 However, the additional warm ischaemia time (WIT) incurred during the DCD donation process has led to a higher reported incidence of complications.3,4

Until recently, all deceased donor liver transplants in Australia were performed with liver allografts retrieved from donors after brain death. However, prior to brain death legislation being established within all jurisdictions, all the early experience in deceased donor organ transplantation was done solely with DCD renal allografts.5 In New South Wales in the 1980s, when transplantation of the other solid organs became a reality with donation after brain death, the practice of DCD renal transplantation all but ceased. All the preliminary experience with liver transplantation in Europe and the United States was with the use of DCD liver allografts,6,7 but the focus switched to donation after brain death after the recognition of brain death as an entity and the enactment of legislation.

With controlled DCD organ donation becoming re-established internationally in the 1990s,8–10 reports then emerged of reasonable outcomes in renal followed by liver and then lung transplantation.11–14 Hence, interest was rekindled in NSW with respect to the DCD pathway for organ donation. This culminated in the release of a jurisdictional policy guideline document by NSW Health in 2007, which also facilitated the development of collaborative multiorgan retrieval surgical protocols between the Australian National Liver Transplantation Unit (ANLTU) and the regional lung transplant unit for DCD donors.15

Methods

Data collected prospectively from 1 July 2007 to 31 December 2010 were analysed. Donor data were obtained from the NSW Organ and Tissue Donation Service, while data on the acceptance and utilisation of the DCD livers for transplantation were obtained from the ANLTU database. Patient and graft outcomes data were also obtained from the ANLTU database, as well as hospital records, with a minimum of 6 months’ recipient follow-up. The study was approved by the South Eastern Sydney and Illawarra Area Health Service and Sydney South West Area Health Service ethics review panels. Statistical analysis was performed with StatsDirect version 2.7.8 (StatsDirect).

ANLTU protocol for liver retrieval from DCD donors

The ANLTU protocol based on international best practice is outlined in the Appendix (online at mja.com.au).15–18

Selection of recipients

Before listing, all potential liver transplant recipients were evaluated by a multidisciplinary liver transplant team. Informed consent about the possibility of the use of a DCD liver allograft was obtained. Patients were accepted onto the liver transplant waiting list in accordance with the Transplantation Society of Australia and New Zealand consensus statement protocols.19 Donor livers were preferentially allocated to recipients in whom the surgical hepatectomy was expected to be straightforward, in an attempt to limit the cold ischaemia time of the donor liver to less than 8 hours.14

Liver transplant process

Liver transplantation was performed with standard operative techniques. A routine postimplantation postreperfusion biopsy sample of the liver was obtained for histological testing, including the grading of steatosis.20 Recipients were managed postoperatively according to ANLTU protocols.

Outcome measures

Data on the donors included donor demographics, underlying cause of death, and the outcomes of the donation and surgical retrieval process. Data on early outcomes (within the first 3 months) were obtained for 14 of the transplanted livers. This included liver allograft function and recipient intraoperative and postoperative course (including biliary and vascular complications). Late outcome data included significant recipient complications as well as recipient and graft survival. Primary non-function and initial poor graft function were defined according to previous publications.20

Results

Acceptance of donor offers for liver transplantation by the transplant team

The number of DCD offers steadily increased over the first 3.5 years as seen in Box 1, with acceptance rates being relatively consistent after the first year. Forty-five of 84 donor offers were not accepted mostly because the donor parameters fell outside of the ANLTU DCD acceptance criteria. The most common reason for non-acceptance was advanced donor age alone (21/45), followed by medical abnormalities combined with advanced donor age (12/45), isolated medical abnormalities (8/45), and other factors (4/45), including organisational logistics and withdrawal of consent for donation. However, thirty-five of these 45 potential DCD donors subsequently provided other organs for transplantation.

Outcomes of the planned DCD surgical retrieval

Surgical retrieval teams travelled to donor hospitals of 39 potential DCD donors during the study period. The liver was successfully retrieved for subsequent transplantation in 15 of these donors. There was no difference in demographics between the donors where the liver was successfully able to be retrieved and the donors where it was not. The median γ-glutamyl transpeptidase level was higher in the donors where the liver was not retrieved (81 IU/L) compared with the donors where it was retrieved (47 IU/L).

For the DCD donors where the liver was successfully retrieved, the location of the potential donor at the time of withdrawal of treatment was within the intensive care unit (ICU) in 12 donors, and in the operating theatre complex for the remaining three. The median time from withdrawal of treatment to declaration of death was 11 minutes (range, 4–19 minutes). The subsequent median WIT was 26 minutes (range, 17–35 minutes). The most common reason for non-retrieval of the liver was that death did not occur within the predetermined time frame of less than 30 minutes (14/24 donors), followed by an abnormality detected in the donor liver (8/24 donors) and unforeseen issues with logistics (2/24 donors). Of the 39 donors where liver retrieval was attempted, 25 provided other organs for transplantation.

Outcomes of DCD liver transplants

There were 13 adult and one paediatric recipients, with a median age of 57 years (range, 4–63 years). One donor liver was not used. All the recipients underwent primary liver transplantation. The paediatric recipient received an urgent transplant with a cutdown of a DCD donor liver. The underlying primary liver diseases were hepatitis C virus (six patients), postalcoholic cirrhosis (four patients), hepatocellular carcinoma (three patients) and fulminant hepatic failure (one patient). At the time when a potential donor was identified, 11 recipients were at home, two were hospitalised and the paediatric recipient was in the ICU. The median cold ischaemia time was 7.7 hours (range, 4.9–9.7 hours). The patient and graft outcomes are shown in Box 2, with a median patient follow-up period of 14.8 months (range, 4–39 months).

The early outcomes were favourable, with no primary non-function or significant initial poor graft function despite the peak serum aspartate aminotransferase levels. This was despite five allografts having moderate isolated microvesicular steatosis on postreperfusion biopsy and one having moderate macrovesicular steatosis. The two vascular complications occurred within the first 3 months, and the hepatic artery stenosis was managed with percutaneous balloon dilatation on two occasions. The one case of early bile leak necessitated reoperation and revision of the biliary anastomosis. An anastomotic stricture of the Roux-en-Y was then diagnosed 3 months later, and corrected through endoscopic management (dilatation). The one patient with an early anastomotic biliary stricture underwent endoscopic retrograde cholangiopancreatography (ERCP) and stenting.

The two patients with anastomotic biliary stricture after 3 months required ERCP and stenting. For both patients, multiple ERCPs and stent changes have been required. In one patient there was also biliary sludge and stone formation. One of these two patients required no further stents after 13 months.

Discussion

Since the re-establishment of organ donation through the DCD pathway in NSW, it has become possible to undertake liver transplantation with liver allografts from DCD donors. However, the relatively high rate of non-acceptance of DCD liver offers during the study period reflects the ANLTU acceptance criteria, which are based on known outcomes of DCD liver transplantation internationally.4,21,22

The high percentage of attempted donor retrievals resulting either in non-retrieval or discarding of the liver is consistent with the nature of the DCD process; additionally, in potential donors, death must occur within a short time frame such that the resulting WIT is less than 30 minutes.3,14 The number of liver allografts discarded at the time of retrieval because of steatosis or perfusion abnormality is also consistent with other reported experience.12

The liver allograft outcomes are consistent with other reports including for the rate of biliary complications.23 Although ischaemic cholangiopathy was not seen in this small series, this may reflect both the short follow-up period and the relatively small number of cases compared with other reported series.3 The increased incidence of vascular complications including hepatic artery stenosis,23 along with the increased requirement for retransplantation, was not seen in this initial experience with DCD liver allografts.22

As the results from the initial experience with the use of liver allografts from DCD donors have proven to be favourable, the ANLTU has made a decision to raise the upper age limit for potential donors to 50 years. As the utility of DCD organ donation remains limited, with only 18% of the donors providing liver allografts, the more common practice of obtaining liver allografts through donation after brain death appears to be a more resource-efficient option.

1 Rates of acceptance of and livers retrieved from donation after cardiac death donors

Year	Donor offers	Non-acceptance of offer for liver transplantation	Livers retrieved for transplantation/potential suitable donors

2007	6	5	1/1
2008	16	8	2/8
2009	26	14	4/12
2010	36	18	8/18

2 Patient and graft outcomes of donation after cardiac death liver transplantation

Outcome

< 3 months
after transplant

> 3 months
after transplant

Early graft function

Median peak AST level (range)

3667 IU/L (919–11 264 IU/L)

Median AST level on Day 3 (range)

371 IU/L (92–1375 IU/L)

Biliary complications

Bile leak

1/14

—

Anastomotic stricture

1/14

2/14

Vascular complications

Portal vein thrombosis

1/14

—

Hepatic artery stenosis

1/14

—

AST = aspartate aminotransferase.

Arthroscopy to treat osteoarthritis of the knee?

To the Editor: In their editorial, Buchbinder and Harris conclude that “The use of arthroscopy for knee osteoarthritis has been allowed to continue, exposing patients to an intervention that is at best ineffective, and at worst, harmful”.1

Each year, HCF funds more than 5000 knee arthroscopies in private hospitals alone, and the primary diagnosis is osteoarthritis (coded
as gonarthrosis [arthrosis of the knee]) in more than 1000 of these procedures.

As such, HCF will endeavour to contribute to the debate by surveying members who have a primary diagnosis of gonarthrosis to collect data on self-assessed benefits of knee arthroscopies. It would be fair to say that the patient’s view of the benefits of the procedure is a leading indicator and should form an integral part of assessing the success of knee arthroscopies for osteoarthritis.

Arthroscopy to treat osteoarthritis of the knee?

In reply: We thank Adams for providing private health insurance data that confirm the continued use
of arthroscopic surgery for patients with osteoarthritis.

We agree that the patient’s view of the benefits of the procedure is fundamental to assessing treatment success. We do not doubt that many patients are happy with the results of arthroscopic knee surgery, but this does not necessarily imply that the surgery has had any specific effect, as satisfaction rates are high after many ineffective placebo treatments. Indeed, high-quality randomised controlled trials have consistently failed to demonstrate clinically relevant self-assessed benefits of arthroscopy compared with sham surgery1 or non-surgical comparators.2–4 Potential risks of arthroscopy are also an important consideration. These include thrombosis, infection, complications of anaesthesia, and increased progression of osteoarthritis and likelihood of joint replacement.5

Satisfaction surveys do not justify the ongoing use of ineffective interventions. While some may find it acceptable to fund care based on perceived effectiveness, in this instance to do so might be doing more harm than good.

Correction: Public reporting of surgeons’ performance

Corrections

Incorrect name of organisation, and further clarification: In “Public reporting of surgeons’ performance” in the 6 May 2013 issue of the Journal (Med J Aust 2013; 198: 399-400), the organisation named as the Australian Vascular Society should have been given as the Australian and New Zealand Society for Vascular Surgery (ANZSVS).

The administrator of the Australasian Vascular Audit, a project of the ANZSVS, has requested that it be noted that carotid endarterectomy is one of a number of procedures audited for outcomes. Other procedures audited are open and endovascular aortic surgery, lower-limb bypass graft surgery and arteriovenous fistulae formation for haemodialysis access (see Beiles CB, Bourke B, Thomson I. Results from the Australasian Vascular Surgical Audit: the inaugural year. ANZ J Surg 2012; 82: 105-111).

General practice: unity in diversity

Most of the world is yet to have access to any effective form of primary care. This is a tragedy created by lack of practitioners and infrastructure, prohibitive fees, and indifference to the plight of the poor. About 300 000 mothers, many of them teenagers, die each year in childbirth for lack of simple primary care in the least economically developed nations (http://www.who.int/mediacentre/factsheets/fs348). But there are encouraging signs. In the 6 years I have been visiting India as part of a review team invited by the Ministry of Health, I have seen maternal mortality reduced by women in remote villages — trained for a couple of days and equipped with no more than a mobile phone and a bicycle — who help pregnant women gain access to safe environments providing postpartum treatment and basic support for mother and child (http://www.nrhm.gov.in/monitoring/international-advisory-panel/minutes-of-the-international-advisory-panel.html). There are now 800 000 of these women working in India. Simple changes that take account of local needs and conditions can have a huge effect.

Australia has about 25 000 general practitioners (http://www.phcris.org.au/fastfacts/fact.php?id=6775) and I marvel at their commitment and creativity as they have adapted to working together in new ways through Divisions of General Practice and now in the more comprehensive Medicare Locals. They are reformatting the way we provide primary care and adapting to the challenge of long-term support for people with serious and continuing illnesses. But there is more to be done. In this issue (doi: 10.5694/mja13.10298), former Royal Australian College of General Practitioners president Claire Jackson reviews progress in response to the National Primary Health Care Strategy. Despite improvements in integrating primary care with other services, achievements in e-health, an increase in general practice registrar positions, and improved coordination of people with complex needs, we still have some way to go with workforce, access to services, integration with the secondary and aged care sectors, and fiscal efficiency.

The diverse roles of general practitioners, as they care for people in individually negotiated ways, are also featured in this issue. Catling-Paull and colleagues bring some data to the much-contested topic of publicly funded homebirth (doi: 10.5694/mja12.11665). Ward and colleagues (doi: 10.5694/mja12.10820) describe the impact of incentives for general practitioners and parents on rates of vaccination, while Kaczmarek and colleagues (doi: 10.5694/mja13.10044) investigate reasons for increased notifications of pertussis in recent years. Zurzolo and colleagues’ finding that many parents of children with food allergies have trouble interpreting food warning labels will not surprise many general practitioners (doi: 10.5694/mja12.11669). Zurynski and Elliott (doi: 10.5694/mja13.10424) highlight the problem of transitional care for older children with rare diseases, and suggest a role for specialist general practice clinics. For clinical fodder, the first article in our new Cardiology series (doi: 10.5694/mja12.11054) revisits the importance of absolute cardiovascular disease risk in primary prevention.

The challenges facing general practice include the sheer diversity of clinical responsibility, making sense of new bureaucratic fantasies and upheavals, adapting to reporting schedules for new services that remind us of the adage “pulling up seedlings to see if the roots are growing”, and distribution of an ageing workforce. Matters of money, lagging fee schedules, and work conditions, such as long hours and lack of locum relief, can cause great stress.

Yet general practitioners remain at their posts. Why? I venture it is because they experience great and appropriate satisfaction from the connections they make with the people who come to them for care, as reflected in Links and Bikou’s tale of coming to understand a patient at the end of life (doi: 10.5694/mja13.10290). Nothing else makes sense as an explanation.

In this issue, we offer articles drawn from a range of disciplines relevant to general practice. This diversity underscores general practice’s central place in the health system of what is, after all, a country blessed with a sophisticated, highly skilled and increasingly integrated health care workforce.

Resistant hypercalcaemia in metastatic parathyroid carcinoma

We report a case of inoperable metastatic parathyroid carcinoma involving life-threatening hypercalcaemia that failed to respond to standard therapy. Our review of available therapeutic modalities showed a paucity of evidence to guide management of patients with these rare tumours. This is the first report of the treatment of malignant hypercalcaemia secondary to parathyroid carcinoma using denosumab.

Clinical record

A 45-year-old man with no significant past medical history presented with a fractured clavicle and evidence of a radiolucent lesion (brown tumour) within the clavicle on x-ray, characteristic for hyperparathyroidism (Box 1). On further investigation, biochemical analysis showing raised serum calcium and parathyroid hormone (PTH) levels confirmed primary hyperparathyroidism. A thyroid ultrasound showed a 39 × 29 × 32 mm mass in the right lobe, suggestive of a tumour. A staging computed tomography (CT) scan of the neck and chest and a bone scan showed bone changes consistent with hyperparathyroidism throughout the axial and appendicular skeleton. There was no evidence of metastatic disease.

The patient underwent a partial parathyroidectomy and right hemithyroidectomy. Histopathological examination confirmed parathyroid carcinoma. The tumour was invading the thyroid gland and there was evidence of lymphovascular invasion. After the operation, the patient developed persistent hypocalcaemia as a consequence of relative hypoparathyroidism. He had an ionised calcium level of 0.68 mmol/L (reference interval [RI], 1.12–1.32 mmol/L). He was diagnosed with hungry bone syndrome, a common problem following parathyroidectomy. This required treatment with high-dose oral calcium (5.4 g daily) and calcitriol (0.25 μg three times daily). Residual tumour was suspected, as his postoperative PTH level remained elevated, at 13.5 pmol/L (RI, 0.7–7.0 pmol/L). A whole-body sestamibi scan (Tc [technetium]-99m MIBI [methoxyisobutylisonitrile]-830MBq) did not confirm this. He remained under surveillance with 4-monthly clinical review with blood tests and annual sestamibi scans.

One year after his surgery, the patient’s PTH level had risen further, to 52.7 pmol/L. His calcium supplements were ceased, as he was hypercalcaemic (serum corrected calcium, 3.11 mmol/L; RI, 2.15–2.60 mmol/L). Despite ceasing supplementation, his hypercalcaemia worsened over 2 months, to a level of 4.15 mmol/L, with a PTH level of 203 pmol/L and creatinine level of 166 μmol/L (RI, 60–110 μmol/L). On clinical examination, he had two small palpable nodules in the right thyroid bed. A CT scan of the neck confirmed a 1.5 × 2.4 × 1.6 cm mass at the site of the previous surgery.

A positron emission tomography scan showed mildly increased fludeoxyglucose activity at the right thyroid bed, consistent with low-volume residual disease, and intense activity in the left retrocrural soft tissue (Box 2). Biopsy specimens were obtained from both sites and were found to be histologically consistent with parathyroid cancer. The patient underwent a planned resection of the known sites of disease, beginning with a right-sided neck dissection. Thereafter, he underwent a left thoracotomy, which showed more extensive pleural deposits and bulky metastatic disease over the diaphragm and crura, only amenable to debulking. A left pleural effusion, non-malignant on cytology, complicated his postoperative recovery.

After surgery, his PTH level remained elevated, at 112 pmol/L. His hypercalcaemia was difficult to manage despite using intravenous hydration and intravenous zoledronic acid (4 mg). His serum calcium level transiently fell from 3.14 mmol/L to 2.77 mmol/L. Cinacalcet 30 mg twice daily was commenced and titrated to a maximal dose of 90 mg four times daily, with no significant improvement in serum calcium levels.

Given the patient’s failure to respond to cinacalcet, in an attempt to reduce tumour size and function, he received two cycles of chemotherapy with a 3-weekly schedule of cyclophosphamide 500 mg/m2 on Day 1, 5-fluorouracil 500 mg/m2 daily on Days 1–4 and dacarbazine 200 mg/m2 on Days 1–4. On presentation for Cycle 2, his calcium level was 4.37 mmol/L and PTH level 335 pmol/L, indicating no biochemical response. He had symptomatic hypercalcaemia with fatigue, polydipsia, anorexia and renal dysfunction for management. He continued to receive weekly bisphosphonate infusions for management, but, despite this, his serum calcium levels did not fall below 3.0 mmol/L.

We trialled denosumab as a means of achieving control of the resistant hypercalcaemia. We used loading doses, aiming to bring about rapid blockade of receptor activator of nuclear factor kappa B (RANK) ligand. The schedule was taken from a Phase II study of giant-cell tumours of bone, in which loading doses of denosumab 120 mg were given subcutaneously on Days 1, 8 and 15, with monthly maintenance therapy thereafter.1 In our patient, after the initial loading dose, the serum-corrected calcium level dropped to 2.66 mmol/L by Day 8. After a second dose, by Day 15 his calcium level was normal, at 2.33 mmol/L. Calculated urinary N-telopeptide levels were monitored to test for suppression of bone turnover. The baseline level was 431 nmol bone collagen equivalents (BCE)/mmol creatinine (RI, < 65 nmol BCE/mmol creatinine); by Day 15 it fell to < 20 nmol BCE/mmol creatinine, indicating rapid suppression of bone turnover. After two loading doses of denosumab, the patient’s serum calcium remained within the normal range for 4 months. His creatinine level also decreased to 102 μmol/L. Further treatment with denosumab was to be guided by the serum calcium level.

Despite control of the calcium level, the patient’s disease progressed and his PTH level rose to a maximum of 433 pmol/L. He became symptomatic, with a recurrent pleural effusion, anorexia and weight loss. Owing to a paucity of data on effective chemotherapeutic agents, the case was discussed at a departmental case review meeting and a regimen of carboplatin and gemcitabine was suggested, given its broad activity and use in carcinoma of unknown primary site.2 The patient completed six cycles of treatment with relative stability of the pleural effusion and retrocrural disease; however, his PTH level did not decrease. His physical condition deteriorated over 6 months after completion of chemotherapy, and he required one further dose of denosumab for calcium control. The patient died 1 year after his presentation with metastatic disease.

Discussion

Parathyroid carcinoma is rare, accounting for less than 1% of cases of hyperparathyroidism. It is associated with a high rate of local and distant recurrence. The mainstay of treatment is surgical resection, even in metastatic disease. The disease is often refractory to medical management, leading to morbidity and mortality due to hypercalcaemia rather than metastatic disease; thus, an aggressive approach to resection has been advocated.3 Managing life-threatening hypercalcaemia and controlling a relatively indolent tumour pose a significant therapeutic challenge in inoperable disease.4

The efficacy of cinacalcet in the management of parathyroid cancer-related hypercalcaemia was assessed in an open-label single-arm study of 29 patients.5 Cinacalcet was commenced at 30 mg twice daily and titrated to 90 mg four times daily or until there was a response in the serum calcium. Over the 16-week study period, 62% of patients responded to therapy; the greatest response was seen in patients with the highest baseline calcium levels. PTH levels were also shown to decrease, but this was not clinically significant.

There is limited evidence for chemotherapy in the management of metastatic parathyroid carcinoma. Success using cyclophosphamide, 5-fluorouracil and dacarbazine has been reported, with normalisation of serum calcium and a complete response in pulmonary metastases after 12 cycles.6 There are several other case reports of agents that have demonstrated some efficacy, including dacarbazine as a single agent, the “MACC” regimen (methotrexate, adriamycin [doxorubicin], cyclophosphamide and CCNU [cyclonexyl-chloroethyl-nitrosourea, or lomustine]) and synthetic oestrogen therapy.4

Few strategies to manage resistant hypercalcaemia in parathyroid carcinoma have been published. One case report described immunisation using the bioactive section of PTH in a patient with resistant hypercalcaemia, with rapid improvement in symptoms and serum calcium. Autoantibodies to parathyroid hormone were identified in the patient’s serum within 4 weeks. The authors theorised that they prevented the binding of PTH to its receptors, thereby improving hypercalcaemia. There was no clinical or radiological evidence of tumour regression after 6 months of therapy.7

Denosumab is a fully human monoclonal IgG2 antibody that has a high affinity and specificity to bind RANK ligand, mimicking the effect of osteoprotegerin, which acts to competitively inhibit the binding of RANK ligand and RANK to inhibit osteoclast activation.8 Denosumab has been compared with zoledronic acid in the prevention of skeletal-related events, such as fractures, in patients with advanced malignancies in two randomised trials. In both, denosumab was associated with higher rates of hypocalcaemia and greater suppression of bone turnover markers.9,10 It is currently licensed in Australia for the treatment of osteoporosis and prevention of skeletal-related events in patients with metastatic breast and prostate cancer with bone metastases.

Preclinical data have shown that telomerase is active in parathyroid cancer cells and could be a potential therapeutic target. An in-vitro study of zidovudine, an antiretroviral agent, investigated its use in inhibiting telomerase. Zidovudine accumulated in malignant parathyroid cells in culture, which induced apoptosis. The effect was not observed in the adenomatous parathyroid cells, suggesting this could be considered as a targeted therapy.11 Studies in humans have not yet been done.

Metastatic parathyroid cancer is a rare malignancy that has a limited evidence base to guide management. We report a patient with refractory hypercalcaemia that dramatically responded to the RANK ligand inhibitor denosumab, with rapid suppression of bone turnover within days. Unlike bisphosphonates, denosumab does not require dose adjustment for renal impairment, which is common in patients with hypercalcaemia. Denosumab offered prolonged calcium control with less frequent administration. The management of parathyroid cancer remains a therapeutic challenge, owing to the minimal evidence for effective systemic therapy in patients with inoperable disease. We propose that denosumab should be considered in the treatment algorithm for patients with hypercalcaemia secondary to parathyroid cancer.

1 Plain x-ray showing brown tumour at the distal end of the left clavicle, characteristic of hyperparathyroidism

2 Positron emission tomography scan showing fludeoxyglucose avidity at the left retrocrural soft tissue, consistent with metastatic disease

The power of one

A core value of clinical medicine is to do the very
best for every single patient. But the principles of evidence-based practice involve developing a management plan that is informed by studies showing the net result of an intervention is balanced towards benefit. Accordingly, some patients will be helped by it, some will experience a null effect, and others will be harmed. How can we reconcile this with the practitioner’s focus on each individual patient’s interests? This question, though not novel, is implicit in many clinical decisions but is rarely articulated.

A current prominent example is the contention over whether to screen men for prostate cancer. Several articles in this issue revisit this question. Martin and colleagues’ analysis (doi: 10.5694/mja12.11597) concludes that prostate-specific antigen screening is likely to only be cost-effective in the < 1% of men classified as very high risk. While this is valuable information, it begs the question of how to identify these high-risk men. From a purely clinical perspective, the balance between the benefits and harms of prostate cancer screening remains unresolved, despite recent large studies, because of disagreement about how to interpret the evidence, say Del Mar and colleagues (doi: 10.5694/mja12.11576). Nevertheless, they offer welcome advice for walking individuals through the maze of uncertainty — but suggest that the conversation only be had at the patient’s request.

In response, Hugosson and Carlsson cite their own study, which showed a survival benefit from screening that was largely offset by an almost equivalent loss of quality-adjusted life-years, mainly from the effects of treatment (doi: 10.5694/mja13.10242).

These can be difficult concepts for doctors, let alone patients. But how do we make a decision about screening for any cancer? According to Rychetnik and colleagues (doi: 10.5694/mja12.11275), our thinking should go beyond benefits, harms and costs, to include social and ethical aspects. They say that “we need to ensure accountability to the citizens whose lives are changed, for better and for worse, by cancer screening programs”.

With regard to vaccination programs — an area of disease prevention with far less ambiguous outcomes — high participation rates are necessary for success. Here, doctors can be confident that the benefits outweigh the harms in spades. Yet when adverse events (rarely) occur, rapid and accurate ascertainment is vital. Clothier and colleagues’ useful study (doi: 10.5694/mja12.11751) will inform postlicensure surveillance after this year’s roll-out of human papillomavirus vaccination for adolescent boys. Using knowledge from the existing program for girls, they estimated the background rates of neurological and allergic events that might be potentially associated with vaccination in boys, to establish
a benchmark for early detection if more than the expected number of events should occur.

Clinical guidelines assist in bringing population-level evidence to bear on individual patient management. However, they are open to interpretation and, even when fully applied to a management plan, not every patient will have the best possible outcome. A father recounts his 29-year-old daughter’s delayed diagnosis of advanced and ultimately fatal colorectal cancer (doi: 10.5694/mja12.11505), and asks the Gastroenterological Society of Australia (GESA) to review its recommendations for investigating iron deficiency anaemia in young women, to advise earlier colonoscopy in severe cases. On behalf of GESA, Sievert and colleagues (doi: 10.5694/mja12.11506) reply that GESA recommendations would advise investigation of this patient’s persistent anaemia, but that the minimal yield from colonoscopy in a very low-risk population for colorectal cancer should be balanced against the procedure’s risks.

“Nightmare” clinical cases like this do not make for comfortable reading. They also give the lie to medicine being solely guided by rigid decision points, protocols and flowcharts. Doctors’ seamless use of the one-on-one relationship with each of their patients — to give context to population-level evidence for the most appropriate management plan — will always be a cornerstone of clinical care. After all, the benefits to patients of having someone versed in the evidence, but relentlessly supporting their interests, far outweigh any harms.