- Email: [email protected]
Monitoring of veno-occlusive disease after bone marrow transplantation by serum aminopropeptide of type III procollagen
Regulation of glucokinase in liver. In: Horecker BL, Stadtman ER, eds. Current topics in cell research. New York: Academic Press, 1976: 1-50. Hers HG, Van Schaftingen E. Fructose 2,6-bisphosphate 2 years after its discovery. Biochem J 1982; 206: 2-12. Karam JH, Salber PR, Forsham PH. Pancreatic hormones and diabetes mellitus in basic and clinical endocrinology. In: Greenspan FS, ed. Basic and clinical endocrinology. New York: Appleton and Lange, 1991: 592-650. Waldhausl M, Bratusch-Marrain P, Gasic S, Korn A, Nowotny P. Insulin production rate, hepatic insulin retention, and splanchnic carbohydrate metabolism after oral glucose ingestion in hyperinsulin anemic type II (non-insulin-dependent) diabetes mellitus. Diabetologia 1982; 23: 6-15. Berger W, Göschke H, Moppert J, Künzli H. Insulin concentrations in
10 Weinhouse S.
11 12
13
14
portal venous and peripheral venous blood in man following administration of glucose, galactose, xylitol and tolbutamide. Harm Metab Res 1973; 5: 4-8. 15 Ishida T, Chap Z, Chou J, et al. Effects of portal and peripheral venous insulin infusion on glucose production and utilization in depancreatized, conscious dogs. Diabetes 1984; 33: 984-90. 16 Heinemann L, Sonnenberg GE, Hohmann A, et al. Pulsatile insulin infusion and glucose-homeostasis in well controlled type I (insulindependent) diabetic patients. J Intern Med 1989; 226: 325-30. 17 Jenkins D, Wolever T, Taylor R, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981; 34: 362-66. 18 Jenkins D, Wolever T, Jenkins A, et al. The glycaemic index of foods tested in diabetic patients: a new basis for carbohydrate exchange favoring the use of legumes. Diabetologia 1983; 24: 257-64.
Monitoring of veno-occlusive disease after bone marrow transplantation by serum aminopropeptide of type III procollagen
Summary of
Differential diagnosis hepatic complications after bone marrow transplantation (BMT) is often difficult. To assess whether serum concentrations of the aminopropeptide of type III procollagen (PIIINP) could facilitate diagnosis, we measured serum PIIINP, corrected for age by conversion to standard deviation scores (SDS), serially after BMT in 27 children. A preliminary study of 11 patients showed that a PNIIINP-SDS cutoff of 8·0 was an optimum for diagnosis of veno-occlusive disease (VOD). PIIINP-SDS was increased above cut-off 1-25 days before the onset of clinical signs in the 12 patients (4 from the preliminary group, the others from a group of 16 studied prospectively) who developed VOD, with subsequent changes in PIIINP-SDS mirroring the course of VOD. By contrast, PIIINP-SDS remained below cut-off in all other patients, including 7 with liver graft-versus-host disease and 3 with drug hepatotoxicity. PIIINP-SDS values greater than 8·0 predict, diagnose, and monitor VOD after BMT.
Introduction Up to 80% of bone
marrow
transplant (BMT) recipients
have hepatic complications, including infections, graftversus-host disease (GvHD), and drug toxicity.1 Cytoreductive drugs used during conditioning therapy can damage hepatic vascular endothelium and lead to venoocclusive disease (VOD), a major cause of mortality after BMT.1Differentiation of the various hepatic complications of BMT is essential for effective management. Serum concentrations of the aminopropeptide of type IIII collagen (PIIINP) are of value in assessing liver disorders.2 Therefore we serially measured serum PIIINP after BMT in children.
Patients and methods Patients 27 children (14 girls, 13 boys; aged 2 months to 16 months, median 5 years) were studied. Indications for BMT included beta-thalassaemia major (8), mucopolysaccharidosis (6), Gaucher’s disease (5), metachromatic leucodystrophy (2), osteopetrosis (2), acute lymphatic leukaemia (1), acute myeloid leukaemia (1), Fanconi’s anaemia (1), and Wiskcott-Aldrich syndrome (1). All patients received cytoreductive conditioning (busulphan and cyclophosphamide)3 with additional conditioning in 4. 16 patients received bone marrow from matched siblings, 7 from matched unrelated donors, and 2 from mismatched donors; 2 patients had autologous BMT. The 25 patients with allogeneic BMT received cyclosporin (to maintain a plasma trough of 40-125 nirnol/L) as GvHD prophylaxis, with additional methotrexate in 13. Complications after BMT included: VOD4 in 12 cases (table); unexplained weight gain of more than 5% from baseline but no
other clinical or biochemical features of VOD in 4; drug hepatotoxicity in 3 (methotrexate in 2, ranitidine in 1); liver, skin, and gut GvHD5 in 7,18, and 10, respectively; infection in 11; haemorrhagic cystitis in 7; hypertension in 4; gastric erosions in 3; acute pancreatitis in 1; exocrine pancreatic insufficiency in 1; and diabetes mellitus in 1. 6 patients died during follow-up, 4 with VOD and multisystem failure (table) and 2 with severe GvHD and overwhelming infection at days 51 and 76 post BMT.
Bone Marrow Transplant Unit (M Eltumi MRCP, Prof J R Hobbs FRCP, C Downie MB), Westminster Children’s Hospital, London; Department of Child Health (P Trivedi MIBiol, P Cheeseman PhD, Prof A P Mowatt FRCP), and Institute of Liver Studies (B Portmann MD), King’s College Hospital, London, UK; and Departments of Clinical Chemistry (J Risteli MD) and Medical Biochemistry (L Risteli MD), University of Oulu, Oulu, Finland
PIIINP and type III
Correspondence to: Dr M Eltumi, Academic Department of Child Health, Chelsea and Westminster Hospital, London SW10 9NH, UK
Serial serum samples (3-36 per patient, median 18, total 473) obtained primarily for management over follow-up of 12-150 days
518
collagen
*1-4 were from first 11 studied, 5-12 from next 16. GD=Gaucher’s disease, HD= = Hurler’s disease, thal=thalassaemia, ost=osteopetrosis, FA=Fanconi’sanaemia, MLD=metachromatic
leukodystrophy. tVOD was diagnosed from clinical criteria4when at least two of the following features were present with no other identifiable cause of liver disease: unexplained weight gain and/or ascites; hepatomegaly and right upper quadrant pain; and jaundice. Diagnosis was confirmed histologically in 5 patients (3 at necropsy) but liver biopsy was considered too hazardous because of thrombocytopenia in other 7. pmpost-mortem, AST=aspartate aminotransferase.
Table : Patients with VOD
(median 72) were analysed for PIIINP with equilibrium radioimmunoassay (Orion Diagnostica).6 Because of the variation in serum
PIIINP during infancy and childhoodabsolute values were corrected for age and sex by conversion to standard deviation scores (SDS): (observed value -mean for age and sex matched controls)=(SD for age and sex matched controls). Paraffin sections (5 11m) from the 5 liver biopsy specimens obtained during VOD and 7 during liver GvHD were stained immunochemically for type IIIcollagen, with an affinity-purified goat antibody to human type III collagen (Southern Biotechnology
with clinical resolution in the 8 who recovered but remained high in the 4 who died. There was no significant difference in peak PIIINP-SDS between those who had more severe VOD (as assessed by number of clinical signs, peak serum
Associates, Alabama). Data
analysis
Data from the first 11 patients suggested that serum PIIINP was significantly greater in those who developed VOD.8 The optimal cut-off for PIIINP-SDS was established with a receiver operator curve and used to calculate the sensitivity, specificity, and "exact" 95% CI ofPIIINP-SDS as a diagnostic marker of VOD in the next 16 patients, who were studied prospectively. Longitudinal changes in PIIINP-SDS were assessed by summary measures,9 and differences in peak values between VOD cases and those with other complications were analysed with U tests.
Results PIIINP as diagnostic marker of VOD In the first 11 patients,8 serum PIIINP-SDS
was significantly greater (p < 0-01) in the 4 who developed VOD (peak 245-1218, median 30-0, table) than in those who did not (1 38-13 22, 5-89). The receiver operator curve from these
data indicated that an increase in PIIINP-SDS above 8-0 in the first 3 weeks post BMT would best diagnose VOD. In the next 16 patients, such increases were found in the 8 who went on to develop VOD and also in 3 who had unexplained weight gain but no other feature of VOD. PIIINP-SDS over 8-0 therefore diagnosed VOD with 100% sensitivity ("exact" 95% CI 63-100 % ) and 63 % specificity (24-91%), with a positive predictive value of 73% (39-94%) and a negative predictive value of 100% (48-100%). Increases in serum PIIINP-SDS in VOD (onset= day - 5 to + 17 post BMT, median + 2) preceded the onset of clinical signs (+ 2 to + 23, +11) and increases in serum bilirubin ( + 2 to + 36, + 15) and AST (+ 1 to + 39, + 16).
Longitudinal changes in
PIIINP
PIIINP-SDS increased progressively in each of the 12
patients as features of VOD developed (figure 1), reaching peaks of 11-2-121-8 (median 28-5, table). PIIINP-SDS fell
Figure 1: Changes In serum PIIINP In patients with VOD and hepatic complications of BMT
other
Dotted !!ne=P!!!NP-SDS8 0.
519
umol/L, median 90 in VOD vs 14-192,98 in the others) and AST (68-1170 IU/L, 162 vs 68-645, 131) but those with VOD had significantly greater peak serum PIIINP-SDS (11.2-121.8,28.5 vs 1.4-13.2,6.9,p < 0.001). Immunostaining of liver biopsy specimens showed accumulation of type IIIcollagen within terminal hepatic venules in VOD, but not in liver GvHD (figure 2). Discussion
Figure 2: Immunoperoxidase staining for type III collagen In biopsy samples
liver
Terminal hepatic venule and perivenular region in liver tissue from (A) patient who died of VOD and multisystem failure at day 43 post BMT (post-mortem biopsy) and (B) patient with liver GvHD at day 38. In (A) lumen was narrowed by subintimal deposition of loosely arranged fibrillar material that stained positively for type III collagen, with staining also at margin of sinusoids, which were dilated around venules and to some distance into acinar zone 3 and 2 of parenchyma. In (B), no positive staining for type III collagen was seen within lumina.
bilirubin, peak AST, or outcome) and those with less severe disease (table). Serum PIIINP-SDS also increased above 80 in the first 3 weeks post BMT in 3 of the 4 patients with unexplained weight gain but no other clinical features of VOD (1 in the first 11 patients and 3 in the next 16) but in these, the increases were less sustained and lower than in VOD, with significantly lower peaks (87-189, median 136, compared with 11’2-121-8, 28-5, p<0001). Serum PIIINP-SDS generally remained below 8 0 in the 12 other patients, 7 of whom had liver GvHD and 3 drug hepatotoxicity (figure 1). 5 of the 69 samples from these patients showed small increases, but, in contrast with those from VOD patients the increases occurred after day 35. Patients with VOD and those with other hepatic complications had similar peaks of serum bilirubin (15-750 520
Serum PIIINP-SDS greater than 8-0 in the first 3 weeks post BMT identified patients who would later develop VOD. Such early identification is useful because prevention or amelioration (eg, with prostaglandins or low-dose heparin10) should begin early. 3 of the 4 patients with unexplained weight gain (but no other clinical or biochemical signs of VOD) also showed increases in serum PIIINPSD S above 8-0 in the first 21 days post BMT which suggests these patients may have had subclinical VOD, but this idea needs further investigation. Distinguishing VOD from other causes of liver damage can be difficult after BMT, especially when thrombocytopenia may pose unacceptable risks for diagnostic percutaneous liver biopsy in the early post-transplant period. Clinical criteria4 are then the only means of diagnosing VOD. Biochemical markers, such as serum concentrations of protein C, protein S, and antithrombin III, may also be useful in the assessment of VOD,10 but have not yet been validated. Our study offers a new approach to the biochemical investigation of VOD by highlighting changes in type III collagen and its metabolites. Such changes include synthesis and accumulation of type III collagen within terminal hepatic venules, which was seen in all 5 liver biopsy samples obtained during VOD. Since PIIINP is released into serum from precursor procollagen molecules during synthesis of type III collagen,6we suggest that the increases in PIIINP-SDS we observed in VOD may, as in other liver disorders,2 reflect at least in part formation and intrahepatic accumulation of type IIIcollagen. However, in VOD, raised serum PIIINP may also be due to defective clearance, because PIIINP is largely cleared by liver sinusoidal endothelial cells" which are likely to be damaged, with vascular endothelial cells, during VOD. In addition, defective hepatic clearance may occur later in VOD when occlusion of the central venules and/or perivenular sinusoidal fibrosis impede hepatic blood flow. By contrast with VOD patients, those with liver GvHD had smaller increases in PIIINP probably because, even though serum AST and bilirubin were increased similarly, intrahepatic type III collagen accumulates less in GvHD. Also clearance of PIIINP is less likely to be impaired in liver GvHD, because bileducts are the main site of damagej in this condition and not, as in VOD, endothelial cells. The small PIIINP increases in liver GvHD are in keeping with our findings in other paediatric liver disorders in which increases are rarely seen, probably because the high concentrations of serum PIIINP during growth mask any small increases due to liver disease.12 The high PIIINP increases in VOD are exceptional in paediatric liver disease, and have only been found in Indian childhood cirrhosis," where they may reflect especially rapid fibrogenesis. We conclude that, in children, a serum PIIINP-SDS of more than 8-0 in the first 3 weeks post BMT predicts VOD, diagnoses the condition with 100% sensitivity, differentiates VOD from other complications of BMT, and
monitors the clinical course. Regular measurement of serumPIIINP-SDS after BMT in children and adults may be important in the identification of patients at risk of VOD and in the assessment of their response to prophylactic
therapy. We thank the staff at the Protein Reference Unit, Westminster Children’s Hospital, for help in the collection and processing of blood samples and Anne Rayner, King’s College Hospital, for help with immunostaining. This study was supported by grants from the COGENT Fund, the Secretary of Science Research (Libya), the Voluntary Research Trust of King’s College Hospital, the Children’s Liver Foundation, the Medical Research Council of the Academy of Finland, and the Finnish Cancer Foundation. We also thank Dr K Hugh-Jones, Dr S Jones, and Dr A Vellodi, Bone Marrow Transplant Unit, Westminster Children’s Hospital, for allowing us to study their patients.
References 1 McDonald GB, Shulman HM, Wolford JL, Spencer GD. Liver disease after human bone marrow transplantation. Semin Liver Dis 1987; 7: 210-29. 2 Plebani M, Burlina A. Biochemical markers of hepatic fibrosis. Clin Biochem 1991; 24: 219-39. 3 Hobbs JR, Hugh-Jones K, Shaw PJ, Downie CJC, Williamson S. Engraftment rates related to busulphan and cyclophosphamide dosages for displacement of bone mnarrow transplantation in 50 children. Bone Marrow Transplant 1986; 1: 201-08. 4 McDonald GB, Sharma P, Matthews DE, Shulman HM, Thomas ED. Veno-occlusive disease of the liver after bone marrow transplantation: diagnosis, incidence and pre-disposing factors. Hepatology 1984; 4: 116-22.
Hershko C, Gale RP. GvHD scoring system for predicting survival and specific mortality in bone marrow transplant recipients. In: Gale RP, Fox CF, eds. Biology of bone marrow transplantation. New York: Academic Press, 1980: 56-59. 6 Risteli J, Niemi S, Trivedi P, Maentausta O, Mowat AP, Risteli L. Rapid equilibrium radioimmunoassay for the aminoterminal propeptide of human type III procollagen. Clin Chem 1988; 34: 715-18. 7 Trivedi P, Hindmarsh P, Risteli J, Risteli L, Mowat AP, Brook CGD. Growth velocity, growth hormone concentrations and serum concentrations of the aminopropeptide of type III procollagen. J Pediatr 1989; 114: 225-30. 8 El-Tumi M, Trivedi P, Jurges E, Vellodi A, Mowat AP, Hobbs JR. Serum hyaluronic acid and type III procollagen aminopropeptide as markers of graft versus host disease and veno-occlusive disease in bone marrow transplant recipients. Bone Marrow Transplant 1990; 5 (suppl 2): 116. 9 Matthews JNS, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990; 300: 230-35. 10 Shulman HM, Hinterberger W. Hepatic veno-occlusive disease—liver toxicity syndrome after bone marrow transplantation. Bone Marrow Transplant 1992; 10: 197-214. 11 Smedsrod B. Aminoterminal propeptide of type III procollagen is cleared from the circulation by receptor mediated endocytosis in liver endothelial cells. Collagen Related Res 1988;; 8: 375-88. 12 Trivedi P, Cheeseman P, Portmann B, Hegarty J, Mowat AP. Variation in serum type III procollagen pepetide in healthy subjects and its comparative value in the assessment of disease activity in children and adults with chronic active hepatitis. Eur J Clin Invest
5
1985; 15: 69-74. 13 Trivedi P, Risteli J, Risteli L, Tanner MS, Bhave S, Pandit AN, Mowat AP. Serum type III procollagen and basement membrane proteins as non-invasive markers of hepatic pathology in Indian childhood cirrhosis. Hepatology 1987; 7: 1249-53.
Damage and regeneration of peripheral treated leprosy
Summary Despite the rapidly falling prevalence of leprosy, the disability and handicap resulting from loss of protective sensation, due to irreversible nerve damage, will remain a huge medical problem for many years. To elucidate the location and consequences of permanent nerve damage in treated leprosy, a prospective study involving nine patients who underwent leg amputation was conducted. Full-length nerves dissected from amputated legs were studied with histological and immunohistochemical methods. Our main findings were that: in both lepromatous and tuberculoid leprosy nerve damage increased distally, culminating in total destruction of dermal nerves and sensory nerve endings; after the therapy-related decrease of inflammation large-scale nerve regeneration took place; and that regenerating axons persisted for decades and in tuberculoid leprosy they might reach the subcutaneous fat of the plantar skin. Armauer Hansen Research Institute (AHRI), All Africa Leprosy Rehabilitation and Training Centre (ALERT), PO Box 1005, Addis Ababa, Ethiopia (T L Miko MD); All Africa Leprosy Rehabilitation and Training Centre, Addis Ababa (C Le Maitre BSc); and Department of Anatomy, Faculty of Medicine, Addis Ababa University, Addis Ababa (Y Kinfu PhD)
Correspondence to: Dr Tivadar L Miko, Department of Pathology, University of Sheffield Medical School, PO Box 596, Sheffield S10 2UL, UK
nerves
in advanced
regeneration was blocked by fibrous nerve endings, and that the theoretical basis of nerve grafting in leprosy is in need of further clarification. In some patients, autologous transplantation of skin flaps, probably irrespective of the duration of loss of sensation, might help in regaining protective sensation. We conclude that
nerve
replacement of the distal-most nerves and
Lancet 1993; 342: 521-25
Introduction In 1992 the World Health Organization (WHO) estimated that leprosy cases had fallen to 55 million worldwide.! The success of the antileprosy multidrug regimen introduced in 1982 enabled the WHO to set the goal of eliminating leprosy by 2000.2 However, the spectacular fall in disease prevalence does not affect the incidence and disability rates.3 Furthermore, Waters4 has emphasised that eradication of leprosy is much more difficult than its elimination. The permanent loss of sensation, even in bacteriologically cured patients, results in hand and foot ulcerations
complicated by septic conditions, multilations,
and
severe
deformities. WHO estimate that at present 2-3 million patients have residual deformity as a result of past leprosy.5 Therefore, after the elimination of leprosy the next step will be to deal with disability and handicap.3 The pathology of large mixed peripheral nerves in leprosy, especially the pathology of leprosy-related permanent damage, is poorly understood. In active leprosy 521
10 Weinhouse S.
11 12
13
14
portal venous and peripheral venous blood in man following administration of glucose, galactose, xylitol and tolbutamide. Harm Metab Res 1973; 5: 4-8. 15 Ishida T, Chap Z, Chou J, et al. Effects of portal and peripheral venous insulin infusion on glucose production and utilization in depancreatized, conscious dogs. Diabetes 1984; 33: 984-90. 16 Heinemann L, Sonnenberg GE, Hohmann A, et al. Pulsatile insulin infusion and glucose-homeostasis in well controlled type I (insulindependent) diabetic patients. J Intern Med 1989; 226: 325-30. 17 Jenkins D, Wolever T, Taylor R, et al. Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr 1981; 34: 362-66. 18 Jenkins D, Wolever T, Jenkins A, et al. The glycaemic index of foods tested in diabetic patients: a new basis for carbohydrate exchange favoring the use of legumes. Diabetologia 1983; 24: 257-64.
Monitoring of veno-occlusive disease after bone marrow transplantation by serum aminopropeptide of type III procollagen
Summary of
Differential diagnosis hepatic complications after bone marrow transplantation (BMT) is often difficult. To assess whether serum concentrations of the aminopropeptide of type III procollagen (PIIINP) could facilitate diagnosis, we measured serum PIIINP, corrected for age by conversion to standard deviation scores (SDS), serially after BMT in 27 children. A preliminary study of 11 patients showed that a PNIIINP-SDS cutoff of 8·0 was an optimum for diagnosis of veno-occlusive disease (VOD). PIIINP-SDS was increased above cut-off 1-25 days before the onset of clinical signs in the 12 patients (4 from the preliminary group, the others from a group of 16 studied prospectively) who developed VOD, with subsequent changes in PIIINP-SDS mirroring the course of VOD. By contrast, PIIINP-SDS remained below cut-off in all other patients, including 7 with liver graft-versus-host disease and 3 with drug hepatotoxicity. PIIINP-SDS values greater than 8·0 predict, diagnose, and monitor VOD after BMT.
Introduction Up to 80% of bone
marrow
transplant (BMT) recipients
have hepatic complications, including infections, graftversus-host disease (GvHD), and drug toxicity.1 Cytoreductive drugs used during conditioning therapy can damage hepatic vascular endothelium and lead to venoocclusive disease (VOD), a major cause of mortality after BMT.1Differentiation of the various hepatic complications of BMT is essential for effective management. Serum concentrations of the aminopropeptide of type IIII collagen (PIIINP) are of value in assessing liver disorders.2 Therefore we serially measured serum PIIINP after BMT in children.
Patients and methods Patients 27 children (14 girls, 13 boys; aged 2 months to 16 months, median 5 years) were studied. Indications for BMT included beta-thalassaemia major (8), mucopolysaccharidosis (6), Gaucher’s disease (5), metachromatic leucodystrophy (2), osteopetrosis (2), acute lymphatic leukaemia (1), acute myeloid leukaemia (1), Fanconi’s anaemia (1), and Wiskcott-Aldrich syndrome (1). All patients received cytoreductive conditioning (busulphan and cyclophosphamide)3 with additional conditioning in 4. 16 patients received bone marrow from matched siblings, 7 from matched unrelated donors, and 2 from mismatched donors; 2 patients had autologous BMT. The 25 patients with allogeneic BMT received cyclosporin (to maintain a plasma trough of 40-125 nirnol/L) as GvHD prophylaxis, with additional methotrexate in 13. Complications after BMT included: VOD4 in 12 cases (table); unexplained weight gain of more than 5% from baseline but no
other clinical or biochemical features of VOD in 4; drug hepatotoxicity in 3 (methotrexate in 2, ranitidine in 1); liver, skin, and gut GvHD5 in 7,18, and 10, respectively; infection in 11; haemorrhagic cystitis in 7; hypertension in 4; gastric erosions in 3; acute pancreatitis in 1; exocrine pancreatic insufficiency in 1; and diabetes mellitus in 1. 6 patients died during follow-up, 4 with VOD and multisystem failure (table) and 2 with severe GvHD and overwhelming infection at days 51 and 76 post BMT.
Bone Marrow Transplant Unit (M Eltumi MRCP, Prof J R Hobbs FRCP, C Downie MB), Westminster Children’s Hospital, London; Department of Child Health (P Trivedi MIBiol, P Cheeseman PhD, Prof A P Mowatt FRCP), and Institute of Liver Studies (B Portmann MD), King’s College Hospital, London, UK; and Departments of Clinical Chemistry (J Risteli MD) and Medical Biochemistry (L Risteli MD), University of Oulu, Oulu, Finland
PIIINP and type III
Correspondence to: Dr M Eltumi, Academic Department of Child Health, Chelsea and Westminster Hospital, London SW10 9NH, UK
Serial serum samples (3-36 per patient, median 18, total 473) obtained primarily for management over follow-up of 12-150 days
518
collagen
*1-4 were from first 11 studied, 5-12 from next 16. GD=Gaucher’s disease, HD= = Hurler’s disease, thal=thalassaemia, ost=osteopetrosis, FA=Fanconi’sanaemia, MLD=metachromatic
leukodystrophy. tVOD was diagnosed from clinical criteria4when at least two of the following features were present with no other identifiable cause of liver disease: unexplained weight gain and/or ascites; hepatomegaly and right upper quadrant pain; and jaundice. Diagnosis was confirmed histologically in 5 patients (3 at necropsy) but liver biopsy was considered too hazardous because of thrombocytopenia in other 7. pmpost-mortem, AST=aspartate aminotransferase.
Table : Patients with VOD
(median 72) were analysed for PIIINP with equilibrium radioimmunoassay (Orion Diagnostica).6 Because of the variation in serum
PIIINP during infancy and childhoodabsolute values were corrected for age and sex by conversion to standard deviation scores (SDS): (observed value -mean for age and sex matched controls)=(SD for age and sex matched controls). Paraffin sections (5 11m) from the 5 liver biopsy specimens obtained during VOD and 7 during liver GvHD were stained immunochemically for type IIIcollagen, with an affinity-purified goat antibody to human type III collagen (Southern Biotechnology
with clinical resolution in the 8 who recovered but remained high in the 4 who died. There was no significant difference in peak PIIINP-SDS between those who had more severe VOD (as assessed by number of clinical signs, peak serum
Associates, Alabama). Data
analysis
Data from the first 11 patients suggested that serum PIIINP was significantly greater in those who developed VOD.8 The optimal cut-off for PIIINP-SDS was established with a receiver operator curve and used to calculate the sensitivity, specificity, and "exact" 95% CI ofPIIINP-SDS as a diagnostic marker of VOD in the next 16 patients, who were studied prospectively. Longitudinal changes in PIIINP-SDS were assessed by summary measures,9 and differences in peak values between VOD cases and those with other complications were analysed with U tests.
Results PIIINP as diagnostic marker of VOD In the first 11 patients,8 serum PIIINP-SDS
was significantly greater (p < 0-01) in the 4 who developed VOD (peak 245-1218, median 30-0, table) than in those who did not (1 38-13 22, 5-89). The receiver operator curve from these
data indicated that an increase in PIIINP-SDS above 8-0 in the first 3 weeks post BMT would best diagnose VOD. In the next 16 patients, such increases were found in the 8 who went on to develop VOD and also in 3 who had unexplained weight gain but no other feature of VOD. PIIINP-SDS over 8-0 therefore diagnosed VOD with 100% sensitivity ("exact" 95% CI 63-100 % ) and 63 % specificity (24-91%), with a positive predictive value of 73% (39-94%) and a negative predictive value of 100% (48-100%). Increases in serum PIIINP-SDS in VOD (onset= day - 5 to + 17 post BMT, median + 2) preceded the onset of clinical signs (+ 2 to + 23, +11) and increases in serum bilirubin ( + 2 to + 36, + 15) and AST (+ 1 to + 39, + 16).
Longitudinal changes in
PIIINP
PIIINP-SDS increased progressively in each of the 12
patients as features of VOD developed (figure 1), reaching peaks of 11-2-121-8 (median 28-5, table). PIIINP-SDS fell
Figure 1: Changes In serum PIIINP In patients with VOD and hepatic complications of BMT
other
Dotted !!ne=P!!!NP-SDS8 0.
519
umol/L, median 90 in VOD vs 14-192,98 in the others) and AST (68-1170 IU/L, 162 vs 68-645, 131) but those with VOD had significantly greater peak serum PIIINP-SDS (11.2-121.8,28.5 vs 1.4-13.2,6.9,p < 0.001). Immunostaining of liver biopsy specimens showed accumulation of type IIIcollagen within terminal hepatic venules in VOD, but not in liver GvHD (figure 2). Discussion
Figure 2: Immunoperoxidase staining for type III collagen In biopsy samples
liver
Terminal hepatic venule and perivenular region in liver tissue from (A) patient who died of VOD and multisystem failure at day 43 post BMT (post-mortem biopsy) and (B) patient with liver GvHD at day 38. In (A) lumen was narrowed by subintimal deposition of loosely arranged fibrillar material that stained positively for type III collagen, with staining also at margin of sinusoids, which were dilated around venules and to some distance into acinar zone 3 and 2 of parenchyma. In (B), no positive staining for type III collagen was seen within lumina.
bilirubin, peak AST, or outcome) and those with less severe disease (table). Serum PIIINP-SDS also increased above 80 in the first 3 weeks post BMT in 3 of the 4 patients with unexplained weight gain but no other clinical features of VOD (1 in the first 11 patients and 3 in the next 16) but in these, the increases were less sustained and lower than in VOD, with significantly lower peaks (87-189, median 136, compared with 11’2-121-8, 28-5, p<0001). Serum PIIINP-SDS generally remained below 8 0 in the 12 other patients, 7 of whom had liver GvHD and 3 drug hepatotoxicity (figure 1). 5 of the 69 samples from these patients showed small increases, but, in contrast with those from VOD patients the increases occurred after day 35. Patients with VOD and those with other hepatic complications had similar peaks of serum bilirubin (15-750 520
Serum PIIINP-SDS greater than 8-0 in the first 3 weeks post BMT identified patients who would later develop VOD. Such early identification is useful because prevention or amelioration (eg, with prostaglandins or low-dose heparin10) should begin early. 3 of the 4 patients with unexplained weight gain (but no other clinical or biochemical signs of VOD) also showed increases in serum PIIINPSD S above 8-0 in the first 21 days post BMT which suggests these patients may have had subclinical VOD, but this idea needs further investigation. Distinguishing VOD from other causes of liver damage can be difficult after BMT, especially when thrombocytopenia may pose unacceptable risks for diagnostic percutaneous liver biopsy in the early post-transplant period. Clinical criteria4 are then the only means of diagnosing VOD. Biochemical markers, such as serum concentrations of protein C, protein S, and antithrombin III, may also be useful in the assessment of VOD,10 but have not yet been validated. Our study offers a new approach to the biochemical investigation of VOD by highlighting changes in type III collagen and its metabolites. Such changes include synthesis and accumulation of type III collagen within terminal hepatic venules, which was seen in all 5 liver biopsy samples obtained during VOD. Since PIIINP is released into serum from precursor procollagen molecules during synthesis of type III collagen,6we suggest that the increases in PIIINP-SDS we observed in VOD may, as in other liver disorders,2 reflect at least in part formation and intrahepatic accumulation of type IIIcollagen. However, in VOD, raised serum PIIINP may also be due to defective clearance, because PIIINP is largely cleared by liver sinusoidal endothelial cells" which are likely to be damaged, with vascular endothelial cells, during VOD. In addition, defective hepatic clearance may occur later in VOD when occlusion of the central venules and/or perivenular sinusoidal fibrosis impede hepatic blood flow. By contrast with VOD patients, those with liver GvHD had smaller increases in PIIINP probably because, even though serum AST and bilirubin were increased similarly, intrahepatic type III collagen accumulates less in GvHD. Also clearance of PIIINP is less likely to be impaired in liver GvHD, because bileducts are the main site of damagej in this condition and not, as in VOD, endothelial cells. The small PIIINP increases in liver GvHD are in keeping with our findings in other paediatric liver disorders in which increases are rarely seen, probably because the high concentrations of serum PIIINP during growth mask any small increases due to liver disease.12 The high PIIINP increases in VOD are exceptional in paediatric liver disease, and have only been found in Indian childhood cirrhosis," where they may reflect especially rapid fibrogenesis. We conclude that, in children, a serum PIIINP-SDS of more than 8-0 in the first 3 weeks post BMT predicts VOD, diagnoses the condition with 100% sensitivity, differentiates VOD from other complications of BMT, and
monitors the clinical course. Regular measurement of serumPIIINP-SDS after BMT in children and adults may be important in the identification of patients at risk of VOD and in the assessment of their response to prophylactic
therapy. We thank the staff at the Protein Reference Unit, Westminster Children’s Hospital, for help in the collection and processing of blood samples and Anne Rayner, King’s College Hospital, for help with immunostaining. This study was supported by grants from the COGENT Fund, the Secretary of Science Research (Libya), the Voluntary Research Trust of King’s College Hospital, the Children’s Liver Foundation, the Medical Research Council of the Academy of Finland, and the Finnish Cancer Foundation. We also thank Dr K Hugh-Jones, Dr S Jones, and Dr A Vellodi, Bone Marrow Transplant Unit, Westminster Children’s Hospital, for allowing us to study their patients.
References 1 McDonald GB, Shulman HM, Wolford JL, Spencer GD. Liver disease after human bone marrow transplantation. Semin Liver Dis 1987; 7: 210-29. 2 Plebani M, Burlina A. Biochemical markers of hepatic fibrosis. Clin Biochem 1991; 24: 219-39. 3 Hobbs JR, Hugh-Jones K, Shaw PJ, Downie CJC, Williamson S. Engraftment rates related to busulphan and cyclophosphamide dosages for displacement of bone mnarrow transplantation in 50 children. Bone Marrow Transplant 1986; 1: 201-08. 4 McDonald GB, Sharma P, Matthews DE, Shulman HM, Thomas ED. Veno-occlusive disease of the liver after bone marrow transplantation: diagnosis, incidence and pre-disposing factors. Hepatology 1984; 4: 116-22.
Hershko C, Gale RP. GvHD scoring system for predicting survival and specific mortality in bone marrow transplant recipients. In: Gale RP, Fox CF, eds. Biology of bone marrow transplantation. New York: Academic Press, 1980: 56-59. 6 Risteli J, Niemi S, Trivedi P, Maentausta O, Mowat AP, Risteli L. Rapid equilibrium radioimmunoassay for the aminoterminal propeptide of human type III procollagen. Clin Chem 1988; 34: 715-18. 7 Trivedi P, Hindmarsh P, Risteli J, Risteli L, Mowat AP, Brook CGD. Growth velocity, growth hormone concentrations and serum concentrations of the aminopropeptide of type III procollagen. J Pediatr 1989; 114: 225-30. 8 El-Tumi M, Trivedi P, Jurges E, Vellodi A, Mowat AP, Hobbs JR. Serum hyaluronic acid and type III procollagen aminopropeptide as markers of graft versus host disease and veno-occlusive disease in bone marrow transplant recipients. Bone Marrow Transplant 1990; 5 (suppl 2): 116. 9 Matthews JNS, Altman DG, Campbell MJ, Royston P. Analysis of serial measurements in medical research. BMJ 1990; 300: 230-35. 10 Shulman HM, Hinterberger W. Hepatic veno-occlusive disease—liver toxicity syndrome after bone marrow transplantation. Bone Marrow Transplant 1992; 10: 197-214. 11 Smedsrod B. Aminoterminal propeptide of type III procollagen is cleared from the circulation by receptor mediated endocytosis in liver endothelial cells. Collagen Related Res 1988;; 8: 375-88. 12 Trivedi P, Cheeseman P, Portmann B, Hegarty J, Mowat AP. Variation in serum type III procollagen pepetide in healthy subjects and its comparative value in the assessment of disease activity in children and adults with chronic active hepatitis. Eur J Clin Invest
5
1985; 15: 69-74. 13 Trivedi P, Risteli J, Risteli L, Tanner MS, Bhave S, Pandit AN, Mowat AP. Serum type III procollagen and basement membrane proteins as non-invasive markers of hepatic pathology in Indian childhood cirrhosis. Hepatology 1987; 7: 1249-53.
Damage and regeneration of peripheral treated leprosy
Summary Despite the rapidly falling prevalence of leprosy, the disability and handicap resulting from loss of protective sensation, due to irreversible nerve damage, will remain a huge medical problem for many years. To elucidate the location and consequences of permanent nerve damage in treated leprosy, a prospective study involving nine patients who underwent leg amputation was conducted. Full-length nerves dissected from amputated legs were studied with histological and immunohistochemical methods. Our main findings were that: in both lepromatous and tuberculoid leprosy nerve damage increased distally, culminating in total destruction of dermal nerves and sensory nerve endings; after the therapy-related decrease of inflammation large-scale nerve regeneration took place; and that regenerating axons persisted for decades and in tuberculoid leprosy they might reach the subcutaneous fat of the plantar skin. Armauer Hansen Research Institute (AHRI), All Africa Leprosy Rehabilitation and Training Centre (ALERT), PO Box 1005, Addis Ababa, Ethiopia (T L Miko MD); All Africa Leprosy Rehabilitation and Training Centre, Addis Ababa (C Le Maitre BSc); and Department of Anatomy, Faculty of Medicine, Addis Ababa University, Addis Ababa (Y Kinfu PhD)
Correspondence to: Dr Tivadar L Miko, Department of Pathology, University of Sheffield Medical School, PO Box 596, Sheffield S10 2UL, UK
nerves
in advanced
regeneration was blocked by fibrous nerve endings, and that the theoretical basis of nerve grafting in leprosy is in need of further clarification. In some patients, autologous transplantation of skin flaps, probably irrespective of the duration of loss of sensation, might help in regaining protective sensation. We conclude that
nerve
replacement of the distal-most nerves and
Lancet 1993; 342: 521-25
Introduction In 1992 the World Health Organization (WHO) estimated that leprosy cases had fallen to 55 million worldwide.! The success of the antileprosy multidrug regimen introduced in 1982 enabled the WHO to set the goal of eliminating leprosy by 2000.2 However, the spectacular fall in disease prevalence does not affect the incidence and disability rates.3 Furthermore, Waters4 has emphasised that eradication of leprosy is much more difficult than its elimination. The permanent loss of sensation, even in bacteriologically cured patients, results in hand and foot ulcerations
complicated by septic conditions, multilations,
and
severe
deformities. WHO estimate that at present 2-3 million patients have residual deformity as a result of past leprosy.5 Therefore, after the elimination of leprosy the next step will be to deal with disability and handicap.3 The pathology of large mixed peripheral nerves in leprosy, especially the pathology of leprosy-related permanent damage, is poorly understood. In active leprosy 521