By Stefanczyk-Sapieha, Lilianna Fainsinger, Robin L
INTRODUCTION Hepatocellular carcinoma (HCC) is an aggressive malignancy frequently associated with chronic liver disease or cirrhosis (1). The incidence and burden of this disease is increasing worldwide, including in Western countries and the United States (2,3). Median survival from the time of diagnosis is reported to range from six to 20 months (1). Surgical and nonsurgical treatment options exist, but are frequently limited by the extent of the underlying liver disease (4). Rare cases of spontaneous remission have been reported in the literature (5-9), however many of these reports have been challenged by physicians caring for HCC patients. This case illustrates some challenges around establishing diagnosis and survival prognostication in HCC, and the negative effect this may have when a palliative diagnosis is assumed.
CASE PRESENTATION
A 56-year-old man was admitted to our palliative care unit (PCU) with a diagnosis of agitated delirium and terminal HCC.
Initial diagnosis of advanced HCC had been made 32 months earlier, when he presented to a different hospital with complaints of vague abdominal pain and a history of weight loss (22 kg) over the previous four months.
His past medical history included chronic alcohol abuse, hepatitis C (possibly linked to previous blood transfusions received for GI bleeds), type 2 diabetes, heavy smoking, depression, and antisocial patterns of behaviour.
During investigation, he was found to have significantly elevated alpha-fetoprotein (AFP) of 2,905 ug/L. He underwent an MRI of the abdomen and was found to have lesions highly suggestive of HCC infiltration, predominantly in the left lateral lobe of the liver. Biopsy of the right lobe was performed to assess the extent of the background liver disease and it revealed cirrhosis (grade 3). Given the severity of underlying liver disease, he was deemed not to be a surgical candidate. Tissue biopsy from the left lobe lesions was not obtained. Diagnosis of terminal HCC was communicated to the patient and family, survival was estimated at approximately six months, and palliative management was proposed. He was discharged home on opioids to control his abdominal pain.
Following this admission, he was managed by his family physician, but had several hospital admissions related to poorly controlled abdominal pain, and problems with opioid toxicity and episodes of delirium.
Admission to the PCU was at the request of the palliative care consult service in another hospital, for management of presumed terminal agitated delirium. The delirium was refractory to treatment; the patient was extremely combative, and required physical restraints and constant supervision. He failed trails of haloperidol, olanzapine, zuclopenthixol, trazadone, quetiapine, methotrimeprazine, and was then started on a subcutaneous infusion of midazolam. His opioid was rotated with a reduced dose, and he was given parenteral hydration.
Careful revision of the available documentation from previous hospitalizations and reassessment of the patient and his blood work, as well as his extended survival without surgery or other treatments, led us to question the accuracy of the terminal cancer diagnosis. At this point the midazolam was discontinued and the agitated behaviour was managed with close patient supervision. He became more alert over the next few days without recurrence of the previous agitated behaviour.
A family conference was called to revise goals of care, given the uncertain circumstances. The family wished to maintain DNR code status as per the patient’s previously expressed wishes. They agreed to pursue active treatments, including intravenous antibiotics and hydration, and further investigations as needed to search for potentially reversible causes of delirium and to revaluate the cancer diagnosis.
In the course of investigations, there was no evidence of brain metastases on MRI of the brain. Minimal small vessel disease and mild atrophy was noted. He was found to have right lower lobe pneumonia and left lower lobe atelectasis on CXR, but no evidence of pulmonary metastases. The pneumonia subsequently resolved.
The MRI of the abdomen was repeated and results were compared with previous findings. A shrunken cirrhotic liver was found, with evidence of portal venous hypertension, including splenomegaly and ascites. The previously identified enhancing areas of the left lobe of the liver, suspicious for HCC, were no longer present. There was an interval decrease in the size of the left lobe.
Blood work revealed relatively normal liver function, with normal INR and transaminases, mild bilirubin and alkaline phosphatase elevation, and a low albumin (23 g /L). The AFP had decreased to 17 ug/L.
A family conference was arranged to discuss these results. The possibility of an initial erroneous diagnosis or a spontaneous remission was discussed. Follow-up imaging was proposed for further cancer surveillance, and discharge planning was initiated. Unfortunately, following resolution of the delirium, the patient was noted to have cognitive difficulties especially with regard to executive functioning. The patient went on pass with the family’s assistance and subsequently did not return to the hospital.
DISCUSSION
Diagnosis of HCC
HCC is the most serious complication of chronic liver disease and is frequently diagnosed late, when the disease is already unbeatable (3,10). Major risk factors for HCC are chronic alcohol abuse, and hepatitis B and C (9-11). Increasing incidence of HCC in Western counties seems to be linked to an increase in hepatitis B and C infections (9,11,12). Chronic alcohol abuse continues to be the most common risk factor in Western countries and has an additive effect with hepatitis C infection, where it doubles the risk of HCC and the cancer tends to develop at an earlier age (13). Some of the other risk factors include diabetes, which also has a synergistic effect with viral hepatitis and alcohol abuse, dietary aflatoxins in mouldy grains, and use of oral contraceptives (14).
Diagnosis of HCC is challenging, as patients usually present with manifestations of the underlying chronic liver disease, and only sudden decompensation of the liver disease may heighten the suspicion of HCC (11). The American Association for the Study of Liver Diseases recommends that patients with underlying chronic liver disease, and rising AFP levels should have a contrast CT or MRI. If found to have large (>2 cm) or multifocal liver lesions with arterial hypervascularity, and increased T2 signal intensity, patients should be diagnosed with HCC (15).
A percutaneous biopsy may not be essential to make the diagnosis of HCC in these cases (15). The biopsy should be performed if appearance of the lesion on the diagnostic imaging is not consistent with HCC, or if surgical resection is contemplated. The biopsy is associated with increased risk of bleeding and seeding of the tumour cells along the needle track (11). According to the European Association for the Study of the Liver Disease (2), diagnosis of HCC can be made if a focal liver lesion is greater than 2 cm in diameter, can be identified in two contrast-enhanced modalities (contrast CT, MRI, or angiography), and displays arterial hyper- vascularization in at least one of the modalities. If such a lesion is identified by only one contrast-enhanced imaging modality, then AFP of more than 400 ng/mL is used to confirm diagnosis (2).
Computed tomographic (CT) imaging has been used to identify liver lesions for the past 20 years and is well established as a reliable modality for detection of larger lesions. However, CT has some limitations: detection of early and small lesions is difficult; subcapsular lesions showing transient focal enhancement may be caused by arterial-portal shunting rather than represent HCC, and may disappear on followup CT; dysplastic nodules show substantial arterial phase enhancement and can simulate HCC (16).
Magnetic resonance imaging (MRI) is the examination of choice for detecting and differentiating liver nodules in cirrhotic liver. It is superior to CT and ultrasound (US), but also has low sensitivity for detecting small lesions (17). Colli et al. (18), in a recent systematic review, compared the accuracy of US, AFP, spiral CT, and MRI, and concluded that MRI was likely the superior modality with pooled sensitivity of 81% and specificity of 85%, but, given the wide range of results, definitive diagnosis may still need to rely on CT- or US-guided tissue biopsy.
An elevated AFP has been used as a serum marker for HCC for years, both in screening and as a diagnostic test (19). An AFP of greater than 200 ug/L has been the frequently used cutoff, considered to be highly specific but not very sensitive (20). An elevated AFP has been used to confirm diagnosis in high-risk patients, and also as a screening strategy together with ultrasound (19), although the utility of AFP as a screening test for HCC appears to be limited (19,20).
Our patient had chronic liver disease with significant weight loss, multiple risk factors for HCC, was found to have an elevated AFP, and had focal lesions suggestive of HCC on the MRI. A biopsy of the right lobe of the liver confirmed advanced cirrhosis; he was not, therefore, deemed to be a surgical candidate. All the above findings were highly suggestive of HCC, but there was no evidence of metastatic disease at the time of diagnosis and there was no histopathological confirmation of HCC. Survival Prognostication in Advanced Cancer and HCC
Survival prognostication remains difficult, even in terminal illness and advanced cancer. It is, however, very important for patients living with life-threatening diseases, their families, clinicians, and health and social services. Illness trajectories tend to be different depending on the nature of the terminal disease. The typical steady decline in the terminal phase of cancer is quite different from a pattern of prolonged slow decline with intermittent deterioration and recovery episodes, as usually observed in chronic organ failure (21).
The literature suggests that, although physicians generally tend to overestimate survival and most of the estimates are incorrect when it comes to a particular patient, clinician prognostication still correlates with actual survival (22,23). Vigano at al. (24) conducted a systematic review of literature on survival prediction and suggest using performance status, presence of cognitive failure, weight loss, anorexia, and dyspnea as independent survival predictors in addition to clinical estimation of survival by the treating physician (24).
Delirium is well recognized to be associated with increased mortality rates and shorter survival in patients with advanced cancer (25,26). Delirium is a common reason for admission to palliative care units, with prevalence approaching 80% to 90% in the hours or days before death (25,26). Although it has a strong association with the dying phase, up to 50% of cases of delirium can be reversible with treatment of the cause (25,26).
Consequences of a Diagnosis of Terminal Cancer
Misdiagnosis of terminal illness appears to be rare. It is difficult to establish the frequency of the misdiagnosis of terminal cancer, since cited estimates range widely, from four in 1,635 admissions (27) to two of 330 referrals (28).
Consequences of misdiagnosis of terminal illness can be detrimental when choices about care are made based on this incorrect information, e.g., DNR status, palliative sedation, or refusal of potentially life-prolonging treatments such as antibiotics for infections.
Terminal diagnosis can cause psychological distress and suffering in terminally ill patients. Sense of hopelessness, profound sadness, depressed mood, sense of worthlessness and helplessness, as well, suicidal ideation and social withdrawal are common in this population (29). Preferences for life sustaining therapies can also be influenced by depression (29).
Patients with a history of drug and/or alcohol abuse are at greater risk for delirium because of somatization and tendency toward chemical coping. Alcohol withdrawal, Korsakoff’s psychosis, or dementia should also be considered in the differential diagnosis. A continuous subcutaneous midazolam drip may be used to induce sedation in patients with refractory agitated delirium (25,26).
The effectiveness of the midazolam in this case may suggest alcohol withdrawal as a possible contributing factor. It seems likely that this patient had an underlying dementia with a superimposed delirium due to reversible causes such as the opioid management, psychoactive drugs, alcohol withdrawal, infection, and dehydration. The review of the underlying presumption that the patient had a terminal diagnosis was key in the decision to discontinue the midazolam and avoid a self-fulfilling prophecy.
Is Spontaneous Remission of HCC Possible?
A literature search of Medline and Pub Med, using key words “regression and spontaneous and hepatocellular”, revealed 115 and 119 titles, respectively. Most citations overlapped; 62 were published case reports of spontaneous regression of HCC. Some were published in languages other than English and, therefore, could not be reviewed.
These case reports describe spontaneous regression of HCC, both partial and complete, with significant shrinkage of the lesions on repeat imaging, and decreasing or normalizing levels of AFP. This includes diagnoses confirmed by histopathological examination of either biopsy samples or surgically excised specimens. In a literature review, Lin et al. (6) cited 27 cases of such spontaneous regression. MezaJunco et al. (30), in a more recent case report with a literature review, cited 61 case reports published between 1982 and September 2006. Given the prevalence of HCC, spontaneous tumour regression represents an extremely rare phenomenon (5-8,30,31). Some of the proposed explanations for this rare phenomenon include spontaneous necrosis of the tumour (6,32) or portal vein thrombosis (5,33). Qualglia et al. (9) refer to cases of large regressing HCC and propose disruption of blood supply, withdrawal of hormonal stimulation, or abstinence from alcohol as possible explanations for the spontaneous regression, in the absence of any medical or surgical interventions (9).
Is it possible that this case represents a rare case of spontaneous regression of HCC? The lack of tissue diagnosis makes this explanation less likely but not impossible. Histopathological reports can also be incorrect, as illustrated by Rees at al. (27), and may need to be challenged if strong doubts arise on clinical grounds. Histological confirmation of diagnosis can be challenging, as it may be difficult to differentiate HCC from dysplastic nodules and large regenerative nodules (9). Nevertheless, our case illustrates the value of biopsy confirmation of malignant disease, which could have prevented some of the subsequent unfortunate events for this patient.
CONCLUSION
This case report illustrates the importance of having confidence that diagnosis of a terminal cancer is correct. If any doubts exist, an appropriate review should be conducted. Aggressive therapy may be appropriate until the diagnosis is confirmed and goals of care are again discussed with the patient and/or family.
This case also illustrates the consequences of a possibly erroneous diagnosis. The burden of suffering imposed on the patient and family included somatization with chemical coping and misuse of prescription opioids, ongoing abuse of alcohol, and aggravation of underlying psychiatric problems. During recurrent hospitalizations for a suicide attempt and delirium, there was an ongoing assumption that the patient was deteriorating from HCC which remained unchallenged once the patient was labelled with terminal cancer. The decision on the PCU to review the diagnosis and discontinue the midazolam when the agitation subsided was central in the positive outcome.
Date received, July 4, 2007; date accepted, January 16, 2008.
REFERENCES
1. Curley SA, Barnett CC, Abdalla EK. Staging and prognostic factors in hepatocellular carcinoma. In: Rose BD (ed). UptoDate (version 15.1) Waltham, Massachusetts: UptoDate, 2007.
2. Talwalkar JA, Gores GJ. Diagnosis and staging of hepatocellular carcinoma. Gastroenterol 2004; 127: S126-S132.
3. Seeff LB. Introduction: burden of hepatocellular carcinoma. Gastroenterol 2004; 127: S1-S4.
4. Hoofnagle JH. Hepatocellular carcinoma: summary and recommendations. Gastroenterol 2004; 127: S319-S323.
5. liai T, Sato Y, Nabatame N, Makino S, Hatkeyama K. Spontaneous complete regression of hepatocellular carcinoma with portal vein tumour thrombus. Hepatogastroenterol 2003; 50(53): 1628-1630.
6. Lin TJ, Liao LY, Lin CL, Shin LS, Chang TA, Tu Hy, et al. Spontaneous regression of hepatocellular carcinoma: a case report and literature review. Hepatogastroenterol 2004; 51(56): 579-582.
7. Ohtani H, Yamazaki O, Matsuyama M, Horii K, Shimizu S, Oka H, et al. Spontaneous regression of hepatocellular carcinoma: report of a case. Surg Today. 2005; 35(12): 1081-1086.
8. Van Halteren HK, Salemans JM, Peters H, Vreugdenhil G, Driessen WM. Spontaneous regression of hepatocellular carcinoma. J Hepatol 1997; 27(1): 211-215.
9. Quaglia A, Bhattacharjya S, Dhillon AP. Limitations of the histological diagnosis and prognostic assessment of hepatocellular carcinoma. Histopathol 2001, 38: 167-174.
10. Schwartz JM, Carithers RL. Clinical features, diagnosis, and screening for primary hepatocellular carcinoma. In: Rose BD (ed). UptoDate (version 15.1) Waltham, Massachusetts: UptoDate, 2007.
11. Davila JA, Morgan RO, Shaib Y, McGlynn KA, EI-Serag HB. Hepatitis C infection and the increasing incidence of hepatocellular carcinoma: a population based study. Gastroenterol 2004; 127: 1372- 1380.
12. Liang TJ, Heller T. Pathogenesis of hepatitis C-associated hepatocellular carcinoma. Gastroenterol 2004; 127: S62-S71.
13. Morgan TR, Mandayam S, Jamal MM. Alcohol and hepatocellular carcinoma. Gastroenterol 2004; 127: S87-S96.
14. Yu MC, Yuan J-M. Environmental factors and risk for hepatocellular carcinoma. Gastroenterol 2004; 127: S72-S78.
15. Bruix J, Sherman M. Management of hepatocellular carcinoma. AASLD practice guideline. Hepatol 2005; 42(5): 1208-1236.
16. Baron RL, Brancatelli G. Computed tomographic imaging of hepatocellular carcinoma. Gastroenterol 2004; 127: S133-S143.
17. Taouli B, Losada M, Holland A, Krinsky G. Magnetic resonance of hepatocellular carcinoma. Gastroenterol 2004; 127: S144-S152.
18. Colli A, Fraquelli M, Casazza G, Massironi S, Colucci A, Conte D, Duca P. Accuracy of ultrasonography, spiral CT, magnetic resonance, and alpha-fetoprotein in diagnosing hepatocellular carcinoma: a systematic review. Am J Gastroenterol 2006; 101: 513- 523.
19. Daniele B, Bencivenga A, Megna AS, Tinessa V. 9 alphafetoprotein and ultrasonography screening for hepatocellular carcinoma. Gastroenterol 2004; 127: S108-S112.
20. Gupta S, Bent S, Kohlwes J. Test characteristics of alpha- fetoprotein for detecting hepatocellular carcinoma in patients with hepatitis C. Ann Intern Med 2003; 139: 46-49.
21. Murray SA, Kendall M, Boyd K, Sheikh A. Illness trajectories and palliative care. BMJ 2005; 330: 1007-1011. 22. Christakis NA, Lamont EB. Extent and determinants of error in doctors’ prognoses in terminally ill patients: prospective cohort study. BMJ 2000; 320:469- 472.
23. Glare P, Virik K, Jones M, Eychmuller, Simes J, Christakis C. A systematic review of physicians’ survival predictions in terminally ill cancer patients. BMJ 2003; 327: 195-198.
24. Vigano A, Dorgan M, Buckingham J, Bruera E, Suarez-Almazor ME. Survival prediction in terminal cancer patients: a systematic review of the medical literature. Palliat Med 2000; 14: 363-374.
25. Centano C, Sanz A, Bruera E. Delirium in advanced cancer patients. Palliat Med 2004; 18 (3): 184-194.
26. Del Fabbro E, Dalai Shalini and Bruera E. Symptom control in palliative care. Part III: dyspnea and delirium. J Palliat Med 2006; 2: 422-436.
27. Rees WD, Dover SB, Low-Beer TS, et al. “Patients with terminal cancer” who have neither terminal illness nor cancer. BMJ 1987; 295: 318-319.
28. Taube AW, Jenkins C, Bruera E. Is a “palliative” patient always a palliative patient? Two case reports. J Pain Symptom Manage 1997; 13: 347-351.
29. Block SD. Psychological issues in end-of-life care. J Palliat Med 2006; 9(3): 751-772.
30. Meza-Junco J, Montano-Loza AJ, Martinez-Benitez B, Cabrera- Aleksandrova T. Spontaneous partial regression of hepatocellular carcinoma in a cirrhotic patient. Ann Hepatol 2007; 6(1): 66-69.
31. Vardhana HG, Panda M. Spontaneous regression of hepatocellular carcinoma: potential promise for the future. South Med J 2007; 100(2): 223-224.
32. Ohta H, Sakamoto Y, Ojima H, Yamada Y, Hibi T, Takahashi Y, et al. Spontaneous regression of hepatocellular carcinoma with complete necrosis: case report. Abdom Imaging 2005; 30 (6): 734- 737.
33. Sakata H, Konishi M, Kinoshita T, Satake M, Moriyama N, Ochiai T. Prognostic factors for hepatocellular carcinoma presenting with macroscopic portal vein tumor thrombus. Hepatogastroenterol 2004; 51(60): 1575-1580.
LILIANNA STEFANCZYK-SAPIEHA, Division of Palliative Care Medicine, University of Alberta, Edmonton, Alberta, ROBIN L. FAINSINGER, Division of Palliative Care Medicine, Department of Oncology, University of Alberta, Edmonton, Alberta, Canada
Stefanczyk-Sapieha, Lilianna, MD
Palliative Care Medicine Resident
University of Alberta
Grey Nuns Hospital
217 – Health Services Centre
1090 You ville Drive West
Edmonton, Alberta
Canada T6L 5X8
Fainsinger, Robin L., MD
Division of Palliative Care Medicine
Grey Nuns Hospital
217 – Health Services Centre
1090 Youville Drive West
Edmonton, Alberta
Canada T6L 5X8
Copyright Center for Bioethics, Clinical Research Institute of Montreal Spring 2008
(c) 2008 Journal of Palliative Care. Provided by ProQuest Information and Learning. All rights Reserved.
Comments