Toxic Hepatitis in Black Patients atal

Document Sample
Toxic Hepatitis in Black Patients atal Powered By Docstoc
					_3      pril 1977                                      EDiC      L        Jo   R       L                                                        571

                     Toxic Hepatitis in Black Patients
                                                                                                   ID             atal
                                        J. W INWRIGHT,                             HO L

                                                                                                         ~ MassM! i _ necrosis
 The clinical and pathological features of toxic centri-
 lobular zonal liver necrosis in Natal Blacks are described.
                                                                                                         l.II   Zonal IlYef   necrosis

 It is suggested that this condition may be caused by the
 toxic action of Callilepis laureola (known to the Zulu
 as ·impila'). This plant and other herbs are used by many
 Blacks for self-medication or are prescribed by herbalists
     and witchdoctors.

      . Afr. med. l., 51, 571 (1977).

Toxic centrilobular zonal liver necro is is the commonest
cause of fatal liver necrosi in the Black population of
   atal. The major, if not the sole cause, is the toxic action
of Callilepis laureola, known to the Zulu as 'impila'. The
illne may last only a few hours, the patient dying in
hypoglycaemic coma; or death may occur within a few
day owing to acute liver failure. A few patient uccumb
from uraemia due to acute renal tubular necro i . In thi                        Fig. 1. Incidence of acute liver neero i .
article we describe the clinical and pathological feature
of toxic hepatitis as manifested in Black patient .              tom relating to liver di ea e are of hort duration, jaun-
                                                                 di e i mild or ab ent, and the hi tory ma reveal the
                          MATERIAL                               administration of a herbal remed . For these rea on we
                                                                 have come to the conclu ion that primary centrilobular
Over the past 17 years the Department of Pathology of            zonal liver necro i i of toxic origin.
the Durban Medical School has carried out some 20000               The number of ca e of centrilobular zonal necro i ha
necropsies on Black subjects over 1 month of age. During         varied from year to year (Fig. 2), with an annual average
this period 383 deaths were found to have been directly          of 14 a es. During the year 1959 - 1963 the necrop y
attributable to acute liver necrosi . In 148 cases the necro i   rate wa particularly high, and thi a count for the
wa of the massive type, while in 235 cases the le ion wa         greater number of a e in thi period. part fr m medico-
a diffu e centrilobu1ar zonal necro is. Zonal necrosis           legal case, the necrop rate in recent year ha been only
may occur in association with evere shock, severe in-            23 o~, and ince many a es of toxic liver ne ro i have
fe tion, or conge tive cardiac failure, but all uch cases        been unsu pected clinically, there i no doubt that many
of econdary zonal necrosis are excluded from thi analy-          death due to thi cau e remain undiagno ed. When thi
I.                                                               department undertook the hi topathological investigation
                                                                 at all ho pital in atal it became evident that many uch
              RESULTS AND DISCUSSIO                              death were erroneou 1y ascribed to cardiac failure ince
                                                                 the appearance of the liver at po tmortem examination
Incidence                                                        simulates the 'nutmeg' liver of congestive cardiac failure.
   The age distribution of patients with massive and zonal
liver necrosis differed appreciably (Fig. 1). The highest                 25
IOcidence of massive necrosis was in the 5 - 14-year-old
group, while about one-third of the cases of zonal necrosi
 ) curred in patient under the age of 5 year. Thi differ-
en e, together with differing clinical features, uggests a
different aetiology. In cases of massive necrosis the ill-
                                                                     ~    15

ness i u ually of I or 2 weeks' duration and j a sociated            "0
                                                                     ~    10
with the marked anorexia and jaundice characteri tic of
  evere viral hepatitis. In cases of zonal necrosis the ymp-         ~
                                                                     z     5

                                            (PATH.), Professor                 1959 '60      '62   '64     '66       '68      '70        'n   '74
                                                                                                       Years 1959 - 1975
                                                                                   Fig. 2.   onuaJ incidence of to ic hepatitis.
572                                        SA     MEDIESE         TYDSKRIF                                    23 April 1977

Clinical Features                                               centrilobular hepatocytes are swollen and show karyolysis.
                                                                In cases of later death from liver failure many of the
   The duration of the illness before admission is usually      necrosed hepatocytes have been reabsorbed and the cen-
very short - in 40% of patients it is less than 1 day and       tral sinusoids have become illstended with blood, and thjs
in 72% less than 4 days. The onset is frequently sudden,        gives the characteristic zonal congestion seen macro-
with upper abdominal pain and vomiting. There is us~ally        scopically. The peripheral hepatocytes contain abundant
a rapidly fatal course after admission, 64% of patients         small fat vacuoles. Inflammatory cell infiltration is mini-
dying within 24 hours and 86% withjn 4 days. Hypoglyc~e­        mal, and mainly mononuclear. In cases of late death from
mia is often demonstrable (81 '10), the blood sugar reachmg     uraemia the liver shows centrilobular collapse of the reti-
very low levels, and fits and coma are common. Jaundice         culum framework (Fig. 5) and regeneration of peripheral
may be present, and disturbed Liver function is demon-          hepatocytes, with occasional ductular proliferation if the
strable where time permits of tests being carried out. Many     necrosis has been unusually severe and has encroached on
patients live long enough for renal failure to develop, and     a portal tract. Mitotic activity is often in evidence. Cho-
72 o~ have a raised blood urea concentration.                   lestasis is seen in bile canaliculi and hepatocytes.
   Careful history-taking may reveal that a herbal remedy          The kidney consistently shows evidence of tubular ne-
has been administered, usually as an enema but sometimes        crosis, and tills is more marked when death has occurred
as a draught. Certain patients, particularly adults, present    only after several days. The necrosis involves both con-
a difficult clinical problem since the features attributable    voluted tubules and loops of Henle, and hyaline, granular
to herbal intoxication are superimposed on the symptoms         and red cell casts are seen in the tubules. Tubular rupture,
of some other illness for willch the patient has taken the      with escape of filtrate into the interstitial tissue, evokes an
 herbal remedy. For example, the initial symptoms might
have suggested respiratory tract infection, when fits or
abdominal symptoms developed suddenly and death occur-

Pathological Features
     ecrospsy findings. Nutrition is often good, especially
in children, and there may be no evidence of other organic
disease. Except in cases in which death has occurred very
rapidly from hypoglycaemia, the liver appears pale and
yellow rather than congested, with a sharply demarcated,
centrilobular, punctate, congested area, which has a speck-
led appearance (Fig. 3). In 50% of cases the liver is reduced
in weight, but this is clearly related to the duration of
the necrosis and results from lysis and absorption of
necrotic hepatocytes. In 70°0 of cases renal weight is
increased, the cortex being pale and swollen. Haemorrhages
are often present in other organs, particularly in the lungs,
less often in the brain, skin and intestine.
   Histopathology. The striking feature in the liver is           Fig. 4. Liver showing centrilobuJar zonal necrosis (phos-
                                                                  photungstic acid haematoxylin X 25).
the sharp demarcation of the centriJobular area of necrosis
(Fig. 4). Tn cases of early death from hypoglycaemia the

                                                                  Fig. 5. Liver showing centrilobuJar collapse of reticulum
       Fig. 3. Liver of patient with toxic hepatitis.             (reticulin X 25).
23 April 1977                               SA     MEDICAL         Jo    R   AL                                         573
inflammatory response. Generally there is interstitial          (umhlandlolhi); Trichilia emelica Vahl (umkhulu); Sle-
oedema and tubulovenous aneurysms are occasionally              phania abyssinica Walp. (umlambane); Adenia gummifera
seen. In cases where death is due to uraemia, tubular re-       (imfulwa); Acacia caffra Willd (umlholo).
generation is usually evident and proliferating epithelial
cells reline necrosed tubules.                                                      CASE REPORT
                                                                One case provided more useful information and is
Sociological Investigation of Toxic Hepatitis                    reported in detail:
  It is known that about 85 % of adult Zulus admitted to            Clinical features: A 6-month-old infant was admitted
 our hospital have consulted witchdoctors before attending      to King Edward VIII Hospital on 29 August 1973 and died
 the hospital, and that consultation of traditional practi-     the same day. There was a history of cough and dyspnoea
 tioners such as inyangas, isangomas and herbalists is wide-    for 4 days followed by vomiting and abdominal distension
 spread. However, patients and relatives are reluctant to       for 1 day. The child had marked tachycardia and appeared
 divulge details of any previous treatment when questioned      acidotic but was not jaundiced. Bronchopneumonia and
 by a doctor, so that, while for many years we have sus-        gastro-enteritis were diagnosed clinically and appropriate
 pected herbal intoxication, we have failed singularly in the   treatment was instituted.
past to discover which toxic plants have been used. Infor-          Pathology: The body was of suboptimal nutritional
 mation and samples from a parent, two of whose children        status, weighing 5320 g. Necropsy revealed bilateral
 died as a result of drinking a 'herbal medicine' in their      bronchopneumonia, a normal heart, and a distended small
 parents' absence, led us to believe imfulwa was the parti-     intestine and colon filled with watery stool. The liver was
cular plant involved. Extensive investigations of extracts      pale but showed pronounced centrilobular zonal" engorge-
 from this plant finally convinced us that imfulwa was          ment. It weighed 175 g (expected weight 200 g). The
 non-toxic, although the 'herbal medicine' supplied in this     cortices of both kidneys were pale and swollen.
case unquestionably caused zonal necrosis in the rat.               Histopathology: The lung revealed a viral pneumonitis
   In 1973 a Zulu social worker was engaged to interview        with secondary bacterial pneumonia. The liver showed
the relatives of patients who appeared to have died from        typical centrilobular zonal necrosis with collapse of the
toxic liver necrosis. In all, 30 cases were investigated but    reticulum, cholestasis, and fatty change in the surviving
repeated visits were often necessary in order to gain the       hepatocytes. The kidney showed eosinophilic proteinous
confidence of relatives. Although the interviewer was a         material in tubular lumina, with degenerating and necrotic
Zulu, she encountered considerable difficulty in per-           tubular epithelium indicative of acute tubular necrosis.
suading people to divulge what medicines had been given            Social worker's report: The mother admitted to having
to the deceased or to supply samples of such remedies.          given the infant an enema which she had prepared from
Indeed, in some cases she deemed it unwise to pursue            a root. While the original extract was not available, a
the investigation in view of the threatening attitudes of       desiccated tuber of the same type was given to the social
relatives. Nevertheless, her questioning did reveal that        worker.
probably in all 30 cases some form of traditional remedy           This root was identified as Callilepis laureola at the
had been given to the deceased. Samples of several tubers          atal Herbarium and aqueous extracts produced liver
and plants were obtained and these were identified bo-          necrosis in rats.
tanically at the atal Herbarium. In some instances rela-
tives were unaware of the nature of the herbal medicine            We wish to thank Mr R. G. Strey, Curator of the Natal
which had been obtained from an inyanga.                        Herbarium, for identification of botanical specimens and for
   The following roots and plants were identified, but were     finding and providing samples of Callitepis tal/reoia, and
                                                                Mrs S. Ngcobo for sociological investigations. This work was
proved to have no hepatotoxic effect on laboratory              supported in part by a grant from the Lady Louisa
animals (Zulu names in brackets): Albizia gummifera             MacGillivray Cancer Research Fund.