Tissue culture of human fetal pancreas. Effects of nicotinamide on insulin production and formation of isletlike cell clusters.

Human fetal pancreas (HFP) is a potential source of beta-cells for transplantation to insulin-dependent diabetic patients. We have previously described a method for tissue culture of HFP that results in the in vitro development of isletlike cell clusters (ICCs) containing a minority of insulin-positive cells. Recently we found that nicotinamide, an inhibitor of poly(ADP-ribose) synthetase, induces an increased islet cell DNA replication both in vivo and in vitro. In this study, this culture technique was used to evaluate the effects of addition of 10 mM nicotinamide on HFP explants cultured in RPMI-1640 medium plus 10% human serum. ICCs developed in 11 of 19 consecutive cultures with nicotinamide increased the yield of ICCs by 40%. Also, the insulin content of ICCs increased approximately 50% with nicotinamide supplementation, although measurements of DNA indicated an unchanged number of cells in each ICC. Neither the rates of insulin release in response to 16.7 mM glucose plus 5 mM theophylline nor the (pro)insulin or total protein biosynthesis rates were affected by nicotinamide addition. The combined results of this study suggest that nicotinamide is useful for stimulating the formation of ICCs from HFP.

Tissue culture of human fetal pancreas. Effects of human serum on development and endocrine function of isletlike cell clusters.

The human fetal pancreas represents a source of insulin-producing beta-cells with a potential for transplantation to diabetic patients. It has previously been shown that such cells can be viably maintained in tissue culture media containing fetal calf serum (FCS) and that these explants continue to synthesize and release insulin. In this study the effects of human serum (HS) on the growth and function of human fetal pancreatic explants have been compared with those of FCS. For this purpose, pancreatic glands, obtained after prostaglandin-induced abortions, were briefly exposed to collagenase, and the digest was cultured in RPMI-1640 medium plus 10% pooled HS or FCS. The outgrowth of isletlike cell clusters (ICCs) was monitored. In 31 of 58 consecutively explanted glands, development of ICCs was observed. In the presence of FCS the outgrowth of ICC took place on top of a fibroblast monocellular cell layer; HS effected less growth of fibroblasts and increased the formation of ICCs about sevenfold compared with explants from the same glands maintained in FCS. However, in the explant cultures with HS, the cell number per ICC, expressed as DNA content, was reduced by 50%. In both FCS and HS the insulin content of the medium showed great variability and progressively declined from day 2 to day 5. The medium glucagon concentration also decreased but not to the same extent as that of insulin. Immunocytochemical-stained ICCs showed insulin- and glucagon-positive cells scattered among most nonstained, presumably nonendocrine cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Tissue culture of human fetal pancreas. Development and function of B-cells in vitro and transplantation of explants to nude mice.

The present study evaluates the development and function of human fetal B-cells in vitro with a view to using such cells in future attempts for transplantation of human fetal pancreas to diabetic patients. A method previously described in our laboratory for preparing islets in vitro from the fetal rat pancreas has been applied and modified for use with human fetal pancreas. Pancreatic glands of different gestational ages were obtained from 37 consecutive prostaglandin-induced abortions. After a mild collagenase treatment, the partially disintegrated tissue was maintained in culture for 7 days in tissue culture medium RPMI 1640 plus 20% fetal calf serum to permit cell attachment and out-growth of endocrine cells. In 17 of the 37 consecutively cultured fetal pancreatic glands, islet-like cell clusters were formed. The 20 remaining glands were lost because of either bacterial contamination or lack of viability already before dissection had occurred. Sections of the newly formed cell clusters revealed well-preserved pancreatic cells showing frequent mitotic figures. The tissue exhibited a high rate of (pro)insulin biosynthesis and a modest insulin response to secretory stimuli, suggesting that the mechanism of glucose regulation by the fetal B-cells is not yet fully developed. Electron micrographs showed a large number of granule-containing cells, some of which were identified as B-cells. In nine cases, harvested cell clusters were implanted beneath the kidney capsule of nude mice. When these animals were killed after 2 mo, seven mice showed a considerable growth of the grafts with numerous islet-like structures containing insulin- and glucagon-positive cells.(ABSTRACT TRUNCATED AT 250 WORDS)

Experimental pancreatic transplantation in diabetes.

Although transplantations of vascularized pancreas in diabetic patients show steadily improving results, the immediate operative risks and life-long immunosuppressive medication involved represent considerable disadvantages. Efforts are being made to develop simpler and safer methods of transplantation with isolated pancreatic islet grafts, e.g., isolated islets, fetal pancreas, or dispersed adult pancreas. Iso-, allo-, and xenografts of such preparations have been shown to reverse diabetes in animals. However, attempts to apply these techniques in clinical practice have remained largely unsuccessful, and major technical advances are needed before success is achieved. Attempts to use whole, segmented, or isolated islets from pancreatic grafts as a cure for diabetes in animals and in diabetic patients are reviewed. The importance to the graft's permanent function, of adequate preparation and storage of the graft, and of beta-cell growth and vascularization are reviewed. Various forms of immunomodulation by pretreatment of grafts in vitro have been employed in animal models of diabetes, but none of these have yet been employed with long-term success in humans. Recurrence of a specific autoimmune response toward the beta-cell in a spontaneously diabetic recipient is a potential mechanism for destruction of transplanted islet tissue.

Tissue culture of human fetal pancreas: growth hormone stimulates the formation and insulin production of islet-like cell clusters.

The human fetal pancreas (HFP) is a potential source of insulin-producing B-cells for transplantation to insulin-dependent diabetic patients. We recently described a technique for culturing HFP tissue in vitro which results in the development of islet-like cell clusters (ICC). These clusters exhibited (pro)insulin biosynthesis and a modest rate of insulin secretion, and immunocytochemical staining indicated the presence of insulin-positive cells in the cell clusters. In this study this technique was used to evaluate the effects of the addition of 1000 micrograms/L GH to HFP cultured in medium RPMI-1640 plus 10% human serum. ICCs developed in 21 of 33 consecutive cultures. GH increased the yield of ICC by 35% compared to explants supplemented with human serum alone. The insulin content of the ICCs also was increased, but the size of individual ICCs was not affected by GH, as reflected by an unchanged DNA content. GH also caused increased insulin release when the ICCs were stimulated with 16.7 mM glucose plus 5 mM theophylline. However, (pro)insulin biosynthesis was not affected by the addition of GH. These results suggest that GH stimulates the formation of both ICCs and insulin production within the explants. These observations are relevant both for the production of human fetal B-cells intended for transplantation into insulin-dependent diabetic patients and for our knowledge of the growth regulation of the HFP B-cell.

Pharmacokinetic interaction between single oral doses of diltiazem and sirolimus in healthy volunteers.

AIM AND BACKGROUND: The pharmacokinetic interaction between sirolimus, a macrolide immunosuppressant metabolized by CYP3A4, and the calcium channel blocker diltiazem was studied in 18 healthy subjects. Several clinically important interactions have previously been reported for other immunosuppressive drugs that are metabolized by the same enzyme and for calcium antagonists. METHODS: Healthy subjects who were 20 to 43 years old participated in an open, three-period, randomized, crossover study of the pharmacokinetics of a single 10-mg oral dose of sirolimus, a single oral 120-mg dose of diltiazem, and the two drugs given together. The three study periods were separated by a 21-day washout phase. RESULTS: The geometric mean (90% confidence interval) whole blood sirolimus area under the plasma concentration time-curve increased 60% (35%-90%), from 736 to 1178 ng x h/mL, and maximum concentration increased 43% (14%-81%), from 67 to 96 ng/mL, with diltiazem coadministration, whereas the mean elimination half-life of sirolimus decreased slightly, from 79 to 67 hours. Apparent oral clearance and volume of distribution of sirolimus decreased with 38% and 45%, respectively, when sirolimus was given with diltiazem. The plasma maximum concentration and area under the plasma concentration-time curve of diltiazem, desacetyldiltiazem, and desmethyldiltiazem were unchanged after coadministration of sirolimus, and no potentiation of the effects of diltiazem on diastolic or systolic blood pressure or on the electrocardiographic parameters was seen. CONCLUSIONS: Single-dose diltiazem coadministration leads to higher sirolimus exposure, presumably by inhibition of the first-pass metabolism of sirolimus. Because of the pronounced intersubject variability in the extent of the sirolimus-diltiazem interaction, whole blood sirolimus concentrations should be monitored closely in patients treated with the two drugs.

Liposomal amphotericin B prevents invasive fungal infections in liver transplant recipients. A randomized, placebo-controlled study.

Eighty-six consecutive liver transplant recipients were prospectively randomized in a double-blind, placebo-controlled antifungal prophylaxis study. Seventy-seven patients received 5 days of prophylaxis starting during the transplantation with either liposomal amphotericin B (AmBisome) 1 mg/kg/day or placebo. Among 40 AmBisome-treated patients, no invasive Candida infection was seen during the first month, compared with 5 invasive Candida albicans infections among 37 control patients (P < 0.05). Furthermore, 1 placebo patient experienced Aspergillus niger pneumonia. Thus, the overall incidence of invasive fungal infections was 0/40 (0%) in the AmBisome group versus 6/37 (16%) in the placebo group (P < 0.01). Patient survival at 30 days was 92% versus 94% for AmBisome- and placebo-treated patients, respectively. One patient experienced backache related to AmBisome infusion. Two patients had transient thrombocytopenia possibly caused by AmBisome treatment. AmBisome was otherwise well tolerated. The total cost for all antifungal drugs used in both groups was equal. However, prophylaxis with AmBisome was $5000 less expensive than treatment of proven invasive fungal infections among placebo patients.

Pretransplant herpesvirus serology and acute graft-versus-host disease.

Logistic regression was used to analyze the influence of pretransplant herpesvirus antibodies, in both patients and donors, on the development of acute graft-versus-host disease in 111 consecutive HLA-identical bone marrow recipients. In bivariate analysis, recipient seropositivity for cytomegalovirus (P = 0.01), donor seropositivity for herpes simplex virus (P = 0.02), and low bone marrow cell dosage (P less than 0.05) were associated with a high incidence of grade II-IV acute GVHD. In multivariate analysis the P values were P less than 0.05 for a positive recipient CMV serology and P = 0.07 for a positive donor HSV serology. Positive serology for 1-2 herpes-viruses among recipients or donors both resulted in a 12% incidence of grade II-IV acute GVHD. Positive serology for 3-4 herpesviruses among patients or donors resulted in an incidence of 32% and 38% of acute GVHD, respectively (P less than 0.05). It is concluded that recipient and donor pretransplant herpesvirus immunity can be used to calculate the risk of moderate-to-severe acute GVHD.

Injection of xenogeneic endocrine pancreatic tissue into the portal vein--effects on coagulation, liver function, and hepatic hemodynamics. A study in the pig-to-dog model.

The safety of injecting discordant xenogeneic fetal endocrine pancreatic tissue into the portal vein was studied in a pig-to-dog system. It was found that minced fetal porcine pancreas and fetal porcine isletlike cell clusters prepared by collagenase digestion and culture could be injected with only minor or no hepatic hemodynamic disturbances. Coagulation studies revealed a small increase in plasma fibrinopeptide A, but this increase could be prevented by heparinization of the recipient. There was no consumption of fibrinogen or platelets. In contrast, injection of minced adult porcine pancreas caused pronounced hepatic hemodynamic changes and marked coagulation abnormalities, indicating consumption coagulopathy. The present finding that fetal porcine pancreas can be injected intraportally without deleterious effects in dogs provides a foundation for the eventual clinical use of such material as treatment for insulin-dependent diabetes mellitus.

Early recovery of lymphocyte response to Candida after bone marrow transplantation in colonized patients.

Blood lymphocyte responses to Candida protein antigen (CP, Candida mannan mitogen (CM) and Staphylococcus aureus protein A (SPA) were followed in 133 bone marrow transplant (BMT) recipients. Lymphocyte proliferative responses to CP and SPA normalized within 6-12 months. The response to CM was only decreased in non-colonized patients during the first 2 months post-BMT and had already returned to normal by 1 month in colonized patients. During the forthcoming years all three lymphocyte stimulatory tests responses reached donor levels. There was a tendency for low lymphocyte reactivity to CP, high reactivity to CM and equal reactivity for SPA among recipients of T cell-depleted marrows as compared to patients with conventional graft-versus-host disease (GVHD) prophylaxis. After BMT, patients who were colonized with Candida recovered and increased their lymphocyte response to CP and CM within 1-3 months, while non-colonized patients required more than 6 months to regain normal stimulatory capacity. Patients who developed grades II-IV acute GVHD had significantly higher pretransplant lymphocyte responses to both CP (p = 0.02) and CM (p = 0.02) than patients with grades 0-I acute GVHD. There was also a trend for patients who later had a confirmed invasive Candida infection to have high responses to CP before BMT (p = 0.07). After BMT, stimulation by CM, CP and SPA was not affected by acute or chronic GVHD or cytomegalovirus. In conclusion, proliferative capacity to Candida is restored early after BMT and superficial fungal colonization seem to be of importance for this maturation.

An increased risk of relapse in cyclosporin-treated compared with methotrexate-treated patients: long-term follow-up of a randomized trial.

Patients with haematological malignancies and HLA-identical marrow donors were randomized to treatment with cyclosporin A (CSA, n = 30) or methotrexate (MTX, n = 29) with a follow-up ranging from 32 to 70 months. The two groups were comparable regarding disease status and age. Acute graft-versus-host disease (GVHD) was similar and the cumulative incidences of chronic GVHD was 42% in both groups. The overall incidence of cytomegalovirus (CMV) infection and other late infections were also the same in the two groups. Interstitial CMV pneumonitis occurred in 13% in the CSA group compared with 32% in the MTX group (ns). The probability of relapse was 42% after 4 years among the CSA patients and was significantly higher than the probability of relapse in the MTX patients which was found to be 10% (p = 0.03). The actuarial survival after 5 years was 53% for the CSA patients and 57% for the MTX patients (ns). The relapse-free survival was 41% and 59%, respectively (ns). There were no differences between the two groups in terms of renal or hepatic function, incidence of cataracts, blood cell counts, bone marrow cellularity or Karnofsky scores at 2 and 4 years after transplantation.

Deep Candida infection in children receiving allogeneic bone marrow transplants: incidence, risk factors and diagnosis.

Fifty-eight children, who received 60 allogeneic bone marrow transplants (BMT), were studied with regard to incidence, risk factors and diagnosis of deep Candida infection (DCI). Serum samples were analysed for the presence of Candida IgA, IgM and IgG antibodies and free C. albicans glucoprotein antigen (Ag). Five children (8.7%) had a confirmed DCI and died before engraftment of the new bone marrow. When four patients with suspected deep Candida infection (SDCI) were included, the incidence was 15.6%. Four of the five children (80%) with DCI had pathological Candida IgM antibody (Ab) titers and/or free C. albicans glucoprotein Ag, 2-50 days before DCI was verified by culture, direct microscopy and/or autopsy. Risk factors, using Fisher's exact test for DCI, included not receiving bone marrow from an HLA-identical sibling donor, having a seropositive Herpes simplex virus (HSV) donor and pathological IgA and/or IgM Ab titers against Candida before BMT. In conclusion, a child with the above-mentioned risk factors, runs a risk of acquiring fatal DCI before engraftment. The institution of systemic antifungal prophylactic treatment may prevent death from DCI. After BMT, serological examinations may be of value in the early detection of DCI.

Variables predicting deep fungal infections in bone marrow transplant recipients.

Deep fungal infections (FI) were diagnosed in 27 out of 209 consecutive bone marrow transplantation (BMT) recipients. Autopsy verified that the incidence of deep FI was 10% and the overall incidence was 13%. Using bivariate logistic regression analysis at the time of BMT, high recipient age (p = 0.003), low bone marrow cell dose (p = 0.007), recipient cytomegalovirus (CMV) seropositivity (p = 0.009) and splenectomy (p = 0.03) were significant risk factors for deep FI. In multivariate analysis, splenectomy (p = 0.008), recipient CMV seropositivity (p = 0.01) and low bone marrow cell dose (p = 0.01) held as significant. At 30 days post-BMT anti-thymocyte globulin treatment (p = 0.0006) and graft-versus-host disease grade II-IV (p = 0.005) were significant risk variables in bivariate logistic regression analysis and Fisher's exact probability test. Patients with these risk factors are candidates for treatment with antifungal drugs when they suffer from leukopenia and unclear fever.

Safety of liposomal amphotericin B (AmBisome) in 187 transplant recipients treated with cyclosporin.

The safety of AmBisome was evaluated in 187 transplant recipients treated for 197 episodes. Patients included 89 bone marrow transplant recipients, 64 liver transplant recipients, 20 renal transplant recipients and 14 recipients of combined organs. AmBisome was instituted for verified invasive fungal infection in 34 cases, suspected invasive fungal infections in 80 cases and as prophylaxis in 83 cases. AmBisome was given for a median of 11 days (range 1-112 days) with a maximum daily dose of 1.49 +/- 0.70 mg/kg/day (mean +/- SD). The total cumulative dose of AmBisome was 1.11 +/- 1.78 g (mean +/- SD). Side-effects definitely attributed to AmBisome therapy included low potassium (n = 3), low back pain (n = 3), dyspnoea (n = 2), allergic rash (n = 1), nausea and vomiting (n = 1), confusion (n = 1), rise in alkaline phosphatase (n = 1) and cholecystitis (n = 1) with an overall incidence of 13 of 197 (7%). AmBisome was discontinued due to side-effects in 6 (3%) of the cases. During AmBisome treatment the mean cyclosporin dose was 9.6 +/- 28.8 mg/kg/day. Compared to pre- and post-AmBisome therapy there was a significantly increased cyclosporin concentration in blood during AmBisome therapy. Side-effects with possible association to AmBisome therapy included low serum potassium (36%), increase in serum creatinine (31%), rise in alkaline phosphatases (26%) and fever (3%). The overall mean increase in serum creatinine was 20%. Other possible side-effects like headache, abdominal pain, rash, rise in bilirubin, cramps and pancreatitis was seen in single patients.(ABSTRACT TRUNCATED AT 250 WORDS)

Allogeneic bone marrow transplantation for leukemia: factors of importance for long-term survival and relapse.

From 1977 until June 1987, 139 patients with leukemia underwent allogeneic bone marrow transplantation at Huddinge Hospital. Among recipients of HLA identical bone marrow the survival plateau for 30 patients with acute myeloid leukemia in first remission was 71% from 3 to 7 years. Children (n = 11) had a survival of 100% compared with 52% (n = 19) in adults (p = 0.01). Among patients with acute lymphoblastic leukemia in first remission (n = 15) the actuarial 5-year survival was 73% compared with 31% for those (n = 25) in second to fourth remissions. The probability of relapse in these two groups was 9% and 45% respectively (p less than 0.05). Patients with chronic myeloid leukemia in first chronic phase (n = 27) had a survival plateau from 1 to 6 years of 65%. In Cox's multivariate regression analysis, improved survival was associated with grade 0-I acute graft-versus-host disease (GVHD) (p less than 0.0001), HLA-matched siblings (p less than 0.001), recipient age less than 17 years (p = 0.02), first remission and first chronic phase (p = 0.03), cytomegalovirus (CMV) seronegative recipients (p = 0.04) and absence of symptomatic CMV infection (p = 0.04). A decreased risk of relapse was seen in patients with first remission or first chronic phase (p less than 0.001) and in patients with asymptomatic CMV infection (p = 0.03). In multivariate analysis the p values were 0.002 and 0.09 for these two groups respectively. In these patients acute GVHD was the major obstacle to successful outcome.

Randomized double-blind study of liposomal amphotericin B (Ambisome) prophylaxis of invasive fungal infections in bone marrow transplant recipients.

Invasive fungal infection is a problem in patients undergoing bone marrow transplantation (BMT). To determine if a liposomal formulation of amphotericin B (Ambisome) is safe and can prevent fungal infection we performed a placebo controlled double-blind randomized prophylactic trial. Study drug was administered from when neutrophil count had decreased to < 0.5 x 10(9)/l and was continued until neutrophils recovered to this level or an infection or toxicity end-point was reached. Thirty-six patients received 1 mg/kg/day of ambisome and 40 patients received placebo daily. There were no statistical differences in characteristics or clinical course between the two groups. Fungal colonization decreased in the ambisome group while it increased in the placebo group. By the end of prophylaxis 8 of 24 (33%) patients receiving ambisome were colonized compared with 18 of 29 (62%) placebo patients (p = 0.05). Five and 7 patients on ambisome or placebo, respectively, were withdrawn due to a presumed fungal infection (NS). There was no statistical reduction of autopsy-proven fungal infection. Proven fungal infection occurred in one patient receiving ambisome (C. guillermondi) compared with three patients receiving placebo (C. guillermondi, 2; C. albicans, 1). Ambisome was well tolerated at the dose of 1 mg/kg/day but in three patients allergic reactions were observed.

Fungal prophylaxis by reduction of fungal colonization by oral administration of bovine anti-Candida antibodies in bone marrow transplant recipients.

Candida overgrowth and invasion constitute a serious threat with a high mortality in BMT recipients. Currently available topical antifungal prophylaxis is largely ineffective, and as resistance to existing, absorbable drugs for systemic use is rapidly developing, new forms of therapy are needed. We investigated the effect of oral treatment of BMT recipients with a bovine immunoglobulin product derived from animals immunized against several Candida species. The natural Candida colonization was first followed in 19 patients to establish the colonization pattern. Half of the patients were found to be colonized prior to transplantation and altogether 72% were colonized at some point during follow-up. Those with a high pre-transplant concentration of Candida in saliva (>100 CFU/ml) remained colonized throughout the BMT treatment period. The therapeutic effect was monitored in two other patient groups. The first group consisted of nine patients, where, due to a low number of primary colonized patients, response in colonized patients was suggestive of a therapeutic effect. In the second group, 10 patients with a high level of colonization (>100 CFU/ml) were given 10 g daily of the product in three divided doses. The results suggest a treatment-related reduction in Candida colonization in a majority (7/10) of patients and one patient became completely negative. As no adverse effects were noted, our findings encourage additional studies in immunocompromised, transplant patients.

Prevention of graft-versus-host disease with T cell depletion or cyclosporin and methotrexate. A randomized trial in adult leukemic marrow recipients.

In adult leukemic marrow recipients of HLA identical sibling marrow, 23 patients were randomized to T cell depletion and 25 received cyclosporin (CSA) and four doses of methotrexate (MTX) to prevent graft-versus-host disease (GVHD). Anti-CD8 and anti-CD6 antibodies plus complement depleted 95.3 +/- 5.8% (mean +/- SE) CD3 cells. All patients engrafted except one who died early. Patients receiving T cell-depleted marrow had a faster time to 0.2 x 10(9) WBC/l (p less than 0.001), but required more erythrocyte units (p = 0.03). Platelet transfusions, infections and time in hospital did not differ. The incidence of grade II-III acute GVHD was 23% following T cell depletion and 12% for those receiving CSA + MTX. Chronic GVHD occurred in 51% and 23% in the two groups, respectively (p = 0.06). Recipients of T cell-depleted marrow who developed grade I-III acute GVHD received more T cells compared to those without acute GVHD (p = 0.02). The major cause of death in both groups was relapse, the cumulative incidence of which, at 4 years, was 39% in the recipients of T cell-depleted marrow and 54% in the CSA + MTX group. The 3-year actuarial leukemia-free survival was 42% and 44% in the two respective groups.

Identification of rare Candida species and other yeasts by polymerase chain reaction and slot blot hybridization.

Invasive candidiasis has become a major cause of morbidity and mortality in immunocompromised hosts. Here we describe a fast and reliable DNA extraction and PCR amplification method in combination with a slot blot hybridization assay. A genus-specific probe was designed that allowed to detect DNA from a broad range of Candida species and 3 other yeasts. In addition, species-specific oligonucleotides for emerging Candida and other yeast species allowed to identify DNA extracted from Candida lusitaniae, Candida humicola, Candida kefyr, Candida inconspicua, Candida solani, Malassezia furfur and Trichosporon cutaneum. A sensitivity of at least 10(1) CFU, corresponding to 100 fg of fungal DNA, was documented for all species-specific probes and the common Candida probe. In addition, the 18S rRNA genes of 7 yeast species (C. humicola, C. kefyr, C. solani, C. inconspicua, C. norvegensis, C. utilis and M. furfur) were completely sequenced. The sequencing primers described bind to highly conserved primer binding sites. Therefore, these primers would allow rapid cycle sequence of additional ribosomal genes throughout the whole kingdom of fungi.

Prophylaxis against fungal infections in transplant recipients: possible approaches.

Although the morbidity and mortality associated with invasive fungal infections in transplant recipients is high, the optimal approach to antifungal prophylaxis is controversial. Most fungal infections occur shortly after the trans- plantation during maximum immunosuppression and are caused by Candida or Aspergillus spp. Nonspecific strategies for prevention do not differ from those used in other patients at risk. They consist mainly of a reduction in risk factors, such as removal of plants from around the patient, separation of the patient from the vicinity of construction sites or improvement in hospital care by isolation and careful nursing of the patient, with strict hygiene. A more controversial issue is primary antifungal chemoprophylaxis, since there are few well designed trials of this intervention, and most of the patients studied have had haematological diseases. Orally administered antifungal drugs that are not absorbed through the gastrointestinal tract have not shown any evidence of a prophylactic effect to date. Controlled trials of systemically administered or orally absorbed drugs in specific transplant recipients have, however, proved effective. In allogeneic and autologous bone marrow transplants, fluconazole 400 mg/day was effective when administered from the conditioning treatment period through the neutropenic period. In liver transplant recipients, either fluconazole 400 mg/day for 10 weeks or liposomal amphotericin B 1 mg/kg/day for 5 days significantly reduced the incidence of invasive fungal infections. However, one must be aware of the risk of fluconazole-resistant fungi and the possibility of selection. In patients with a history of previous fungal infection, secondary prophylaxis may be of value, although data are limited. For recipients of transplants other than bone marrow or liver, there are insufficient data to recommend general prophylaxis.