Nephrol Dial Transplant (2000) 15: 285 285

of extrapulmonary involvement of TB, especially in lymph rarely involved by atherosclerosis, and with relatively pre-
served function at old age. Therefore, in adults, the idea of nodes [57] has been reported for uraemic patients. The
clinical course and presentation are non-specic and often using marginal donors for liver transplantation has come
into prominence over the last decade [36]. Usually, these mimic that of underlying chronic renal failure. In addition,
a positive skin test, acid-fast stain, or culture for TB are are elderly patients with medical problems, thought originally
to be unsuitable for organ harvesting, but who are now felt rarely found [5], making the diagnosis dicult. In some
to be acceptable for liver transplantation. But what of cases, the diagnosis can only be conrmed by a successful
chronically dialysed patients? Could the use of liver allografts anti-TB therapeutic trial [5].
from dialysed patients be safe? The case presented here may, Our patient had mild peripheral eosinophilia before CAPD
in a small way, answer this question. therapy, which may be the result of early, occult tuberculous
lymphadenitis [8]. After initiation of CAPD, concomitant
Case: A 60-year-old woman was admitted to hospital with a
mild peripheral eosinophilia and severe peritoneal uid eos-
massive intracerebral haemorrhage. Relevant past history
inophilia persisted over 5 months, even after withdrawal of
included hypertension and end-stage renal failure, both dia-
CAPD. The eosinophilia subsided after 2 weeks of anti-TB
gnosed in September 1993. The patient did not have ADPKD
treatment. The only explanation for the above event was
or known AV malformations. Chronic haemodialysis was
that the peritoneal uid eosinophilia resulted from tubercu-
commenced in October 1993. During this period, a vague
lous lymphadenitis-related peripheral eosinophilia, but not
history of alcoholic abuse was obtained from the family.
vice versa. To our knowledge, this is the rst reported case
Mild elevations of serum liver enzymes were noted (bilirubin
of TB-related peritoneal uid eosinophilia.
0.7 mg/dl; SGOT 5060 U/l; SGPT 35-50 U/l; normal alkal-
In conclusion, in the early stage of initiation of dialysis
ine phosphate level ). All acute viral hepatitis serology (HAV
therapy (usually within the rst year), a high index of
IgM, HbsAg, PCR HCV) was negative. By March 1994, all
suspicion for an aggressive evaluation of TB are necessary if
liver function tests were normal, and they remained so for
uraemic patients present with an unusual and unexplained
the next 4 years.
clinical course. With early diagnosis and adequate treatment,
From 1994 to 1995, symptomatic anaemia necessitated
the outcome of uraemia-related TB is positive [57].
transfusions of 20 units of packed red blood cells.
However, the emergence of multiple drug-resistant TB due
Recombinant human erythropoietin was started in 1996 and
to insucient treatment stemming from patient non-
in January 1997 intravenous iron was added. Serum ferritin
compliance and the high relapse rate of TB associated with
levels, which had peaked at 950 ng/ml in April 1996, dropped
short-term therapy, supports the use of prolonged treatment
to 470 ng/ml by July 1997.
with a combination of multiple drugs for at least 6 months
On admission the patient was intubated and treated sup-
and patient education.
portively with uids and dopamine (5 mg/kg/min). Repeated
Division of Nephrology Sun-Chieh Hsu
liver function tests were normal and viral hepatitis serology
Department of Internal Medicine Rong-Ru Lan
was negative. Brain death was pronounced 18 h after admis-
National Cheng Kung University Chin-Chung Tseng
sion, and immediately thereafter family consent for liver
Hospital Ching-Te Lai
harvesting was obtained. Because of the patients past history,
Tainan Jeng-Jong Huang
a liver biopsy was performed during hepatectomy. On frozen
Taiwan
section, normal liver architecture, but with diuse haemosid-
ROC
erosis in iron-stained Kuppfer cells and hepatocytes, was
seen. This was felt to be secondary to previous blood 1. Chan MK, Chow L, Lam SS et al. Peritoneal eosinophilia in
patients on continuous ambulatory peritoneal dialysis. A pro- transfusions and iron supplementation, and did not serve as
spective study. Am J Kidney Dis 1988; 11: 180183
a contraindication for transplantation.
2. Ejaz AA. Peritoneal uid eosinophilia. Nephrol Dial Transplant
The recipient was a 57-year-old man with end-stage liver
1998; 13: 24632464
failure, secondary to active hepatitis B. Cold ischaemic time
3. Ejaz AA, Fitzpatrick PM, Durkin AJ et al. Pathophysiology of
was 6 h. N-acetylcysteine was started immediately after reper-
peritoneal uid eosinophilia in peritoneal dialysis patients.
fusion and continued for 24 h. The graft showed immediate
Nephron 1999; 81: 125130
function, with adequate bile production and spontaneous
4. Chandran PK, Humayun HM, Dangirdas JT et al. Blood eosino-
correction of coagulopathy. Peak aminotransferase levels
philia in patients undergoing maintenance peritoneal dialysis.
were below 1000 U/l. Mild cellular rejection was treated Arch Intern Med 1985; 145: 114116
5. Belcon MC, Smith EKM, Kahana LM et al. Tuberculosis in successfully at day 14. One month after transplantation the
dialysis patients. Clin Nephrol 1982; 17: 1418 patient was discharged from hospital. Today, at 9 months
6. Vas SI. Renaissance of tuberculosis in the 1990s: Lessons for the
post-transplantation, the patient maintains normal graft
nephrologists. Perit Dial Int 1994; 14: 209214
function.
7. Ellard GA. Chemotherapy of tuberculosis for patients with renal
impairment. Nephron 1993; 64: 169181
Comment: The few absolute contraindications for liver trans-
8. Flores M, Merino-Angulce J, Tanago JG et al. Late generalized
plantation include severe macrosteatosis, liver necrosis or
tuberculosis and eosinophilia. Arch Intern Med 1983; 143: 182
hepatitis in the donor liver [7]. In 1991, Belzers group
showed that if donor livers with biopsy ndings of severe
Procurement of a cadaveric liver transplant from a
fatty liver inltration, hydropic degeneration or centrilobular
chronically haemodialysed patient
necrosis would not have been transplanted, then primary
nonfunction of the transplant could have been reduced from
11% to 5% [7]. Other donor risk factors may also be Sir,
Organ shortage has become the most vexing problem in liver associated with an increased risk for primary nonfunction:
advanced age, obesity, acute infectious problems, hypoten- transplantation, with 1025% of patients dying while
awaiting transplantation [1]. Therefore, new strategies have sion or hypoxia. However, each of these factors, and even
the presence of cardiovascular disease or chronic renal failure been introduced in an attempt to narrow the gap between
demand and supply of organs for such transplantations [2]. are not absolute reasons to discard these so-called marginal
liver donors [5]. As early as 1987, Makowka et al. showed The liver is a privileged organ, with a dual blood supply,
Nephrol Dial Transplant (2000) 15: 286 286
2. Malago M, Rogiers X, Broelsch CE. Liver splitting and living that such marginal donors, when compared to acceptable
donor techniques. Br Med Bull 1997; 53: 860867
donors, had no increase in primary graft nonfunction and
3. Makowka L, Gordon RD, Todo G et al. Analysis of donor
no overall dierence in 2-year patient survival [3].
criteria for the prediction of outcome in clinical liver transplanta-
Up to 10% of dialysed patients die from acute cerebrovas-
tion. Transplant Proc 1987; 19: 23782382
cular diseasemajor ischaemic strokes and intracranial haem-
4. Klintmalm GBG. The liver donor: special considerations.
orrhages [8]. These patients should be considered as potential
Transplant Proc 1988; 20 [Suppl 7]: 911
liver donors, especially if they are free of known liver disease.
5. Mor E, Klintmalm GB, Gonwa TA et al. The use of marginal
A full viral hepatitis screen should be performed immediately.
donors for liver transplantation. Transplantation 1992; 53:
383386 These patients should be treated aggressively in the rst few
6. Mirza DF, Gunson BK, DaSilva RF et al. Policies in Europe on
hours after suering a major cerebrovascular accident. Blood
marginal quality donor livers. Lancet 1994; 334: 14801483
pressure should be maintained with blood products, colloids,
7. DAlessandro AM, Kalyoglu M, Sollinger HW et al. The
and low-dose dopamine. High-dose dopamine is to be
predictive value of donor liver biopsies for the development
avoided as it decreases portal venous blood ow [4].
of primary nonfunction after orthotopic liver transplantation.
Hypervolaemia should also be avoided in order to prevent
Transplantation 1991; 51: 157163
acute liver oedema [4]. If dialysis is indicated, for whatever
8. Kaufman AM, Levin NW. Morbidity and mortality in haemodia-
reason, then it should be performed. Finally, in any dialysed
lysis. In Massry SG, Glassock RJ eds. Textbook of Nephrology,
patient, it would be wise to perform a liver biopsy at 3rd edition, 1995; Williams and Wilkins, Baltimore
9. Kakizoe S, Yanaga K, Starzl TE et al. Frozen section of the liver harvesting.
biopsy for the evaluation of liver allografts. Transplant Proc 1990; Pretransplant liver biopsies have become an accepted part
22: 416417
of donor hepatectomies [7,9]. Acceptable cold ischaemic time
10. Ploeg RJ, DAlessandro AM, Knechtle SJ et al. Risk factors for
for the liver has now reached 12 h. Therefore, when war-
primary dysfunction after liver transplantation-a multivariate
ranted, frozen section of any liver is feasable [7]. Indeed,
analysis. Transplantation 1993; 55: 807813
discarding unacceptable livers, after biopsy, is a far safer
11. Ali M, Fayemi AO, Rigolosi R et al. Haemosiderosis in haemo-
option for the patient than transplantation of such a liver,
dialysis patients. An Autopsy study of 50 cases. JAMA 1980;
as it may prevent the catastrophic possibility of primary
244: 343345
nonfunction [7,9,10]. In 1990 Kakizoe et al. described their
12. Fleming LW, Hopwood D, Shepherd AN et al. Hepatic iron in
experience with 38 frozen section liver biopsies in donors. dialysed patients given iron dextran. J Clin Pathol 1990; 43:
119124 Twenty seven livers were subsequently not transplanted [9].
Liver biopsy in this patient revealed hepatic hemosiderosis.
This histopathological nding is not a contraindication to
Monitoring sialyl Lewis x (CD15s) on peripheral
liver harvesting. As early as 1980, Ali et al. demonstrated
lymphocytes for the diagnosis of acute rejection
that nearly 50% of his studied dialysed patients had signic-
ant hepatic haemosiderosis, but with only a minor degree of
Sir, liver brosis [11]. All ve patients with serum ferritin levels
To prevent graft loss, many investigators look for markers of over 1000 ng/ml had massive liver iron deposits. In 1990,
which are more helpful than serum creatinine for early Fleming et al. showed that all dialysed patients who had
recognition of rejection. Biopsy remains the only means to received long-term parenteral iron dextran, had hepatic iron
denitely establish the diagnosis, but there is a need for non- deposition [12]. Again, the risk of liver damage from the
invasive methods. iron deposition was low [12].
We assessed the expression of the sialyl Lewis x (CD15s) In conclusion, this brief communication shows, for the
antigen on peripheral lymphocytes, performing ow cytome- rst time, that using a heart-beating, liver allograft from a
try using monoclonal antibody (2H5), in 17 patients with chronically dialysed patient is possible. With the ever-growing
rejection, 23 patients without rejection after renal trans- shortage of organs available for transplantation, there is
plantation, and 18 healthy volunteers. ample reason for the clinical nephrologist to regard the
CD15s is a ligand for selectins CD62E and CD62P on dialysis patient, at the proper time, as a potential organ
activated endothelial cells and platelets [1]. Kannagi et al. donor. Even if dialysis patients are to remain marginal
donors, this in itself is not a reason to exclude them from [2] have developed a monoclonal antibody (clone: 2H5),
entering the pool of possible organ donors. For the patient which recognizes a CD15s epitope on activated lymphocytes
suering from severe and irreversible liver failure, receiving that is not recognized by classical antibodies.
a good liver transplantation is life-saving, even if it comes We used clone 2H5 to examine CD15s antigen expression
from a dialysis patient. on activated lymphocytes. CD15s antigen was strongly
expressed on the peripheral lymphocytes of all patients with
Acknowledgements. Our extremely sincere thanks to Ronit Reches
rejection (n=17), but was only weakly expressed in the other
RN, Transplantation Coordinator, for the Rabin Medical Center
groups. These ndings indicated that, in the patients with
(Campus Golda). Without her devotion and her hard work this
elevated serum creatinine, CD15s might be a helpful marker
transplantation would not have been possible.
for pinpointing the cause of creatinine elevation. Particularly
Department of Nephrology Asher Korzets
in patients with long-term renal dysfunction owing to drug
Rabin Medical Center (Campus Eytan Mor1
nephropathy, ATN, infection and arterial calcication,
Golda) Avri Chagnac
CD15s monitoring may be useful for the dierential diagnosis
1Liver Transplantation Unit Yaacov Ori
of rejection or other causes after renal transplantation.
Rabin Medical Center (Campus Nathan Bar Nathan1
Biopsy was performed in 37 patients who were suspected
Beilinson) Uzi Gafter
to have rejection. Among them, more-than-moderate cell
Petach Tikva
inltration was observed in all of the 17 patients with strong
Israel
expression of CD15s. The expression of CD15s on the
peripheral lymphocytes was almost consistently accompanied
1. Alexander JW, Vaughn WK. The use of marginal donors
by pathological ndings in the rejected kidneys.
for transplantation: inuence of donor age on outcome.
Transplantation 1991; 51: 135137 In all patients given steroid pulse therapy, CD15s antigen