Severe cytomegalovirus enterocolitis after standard chemotherapy for non-Hodgkin's lymphoma

Reports of cytomegalovirus (CMV) colitis mainly concern patients with immunocompromisation resulting from, among others, HIV infection, allogeneic bone marrow transplantation and solid organ transplantation. CMV colitis rarely occurs during standard chemotherapy for non-Hodgkin's lymphoma (NHL). An unusual case of CMV enterocolitis in a 62-year-old patient is reported. After a first course of salvage chemotherapy for NHL, diffuse erosions and sloughing mucosa were seen throughout the large bowel. The final diagnosis was based on histological findings. Although ganciclovir and foscarnet are effective for CMV viremia, their use in the treatment of severe diarrhea in our patient did not result in improvement for one week, whereas concomitant use of octreotide led to rapid improvement. Octreotide may therefore be an effective agent for severe colitis.

Risk for cytomegalovirus infection following reduced intensity allogeneic stem cell transplantation

Preliminary data suggest a faster immune recovery following non-myeloablative stem cell transplantation because of the persistence of recipient T cells, but the real impact on post-transplant infectious complications remains unknown. We retrospectively analysed the incidence of cytomegalovirus (CMV) infection in twenty patients following reduced intensity conditioning with busulfan/fludarabine+/-thiotepa and post-transplant immunosuppression with cyclosporine A/mycophenolate mofetil. Results were compared with 20 patients receiving myeloablative transplants during the same time period and who were matched for CMV risk group and for donor origin. The cumulative incidence of CMV infection following reduced intensity vs. myeloablative transplants was 60.4% vs. 40.0%, respectively (p value 0.1, log rank test). The risk for CMV infection in both cohorts was increased after in vivo T cell depletion with antithymocyte globulin (75% and 60%, respectively). Acute GVHD preceded the diagnosis of CMV infection by a median of 25 (range, 9-61) days following reduced intensity transplants and a median of 14 (range, 10-34) days in myeloablative transplants. Recurrent CMV infections were observed only in patients receiving reduced intensity transplants. Using multivariate analysis only reduced intensity transplantation and in vivo T cell depletion had a significant impact on the risk of CMV infection. In our series the incidence for CMV infection following reduced intensity transplants seems to be increased as compared with risk-matched myeloablative transplants. When adding anti-T cell antibodies to the conditioning regimen, the risk for CMV infection increases by up to 75%. Thorough studies of the risk of post-transplant viral infection are necessary to optimize surveillance as well as pre-emptive and/or prophylactic treatment strategies in the non-myeloablative transplantation setting.

Cytomegalovirus-specific CD4+ and CD8+ T-cells follow a similar reconstitution pattern after allogeneic stem cell transplantation

Cytomegalovirus (CMV) is a common herpes virus that can cause significant morbidity and mortality in immunocompromised individuals, particularly those undergoing allogeneic stem cell transplantation (SCT) for hematological malignancies. Recent studies have examined the kinetics of CMV-specific CD8+ T-cell reconstitution after SCT transplantation and have found virus-specific cytotoxic T-lymphocyte regeneration to be dependent on CMV serologic status and CMV reactivation events. However, the reconstitution kinetics of CMV-specific CD4+ T-cells under these same circumstances were not addressed. In this study, we used HLA class I peptide tetramer for CMV pp65 and cytokine flow cytometry to follow the reconstitution of both CD4+ and CD8+ CMV-specific T-cells after allogeneic SCT. We found that following SCT in which both donors and recipients are CMV seropositive, virus-specific CD4+ T-helper cells show the same reconstitution kinetics as CD8+ cytotoxic T-cells. Following CMV reactivation, a synchronous but temporary increase in both CD4+ and CD8+ CMV-specific lymphocytes occurs. The pattern repeats itself after subsequent episodes of CMV reactivation. These data imply that both CD4+ and CD8+ lymphocytes are necessary for an efficient immune response to CMV and suggest that CD4+ and CD8+ CMV-specific T-cells are required for the complete restoration of CMV immunity. These findings may have important implications in the development of CMV-specific adoptive immunotherapy strategies.

Anti-CD40 therapy extends renal allograft survival in rhesus macaques

BACKGROUND

Organ transplant recipients currently require lifetime immunosuppressive therapy, with its accompanying side effects. Biological agents that block T-cell costimulatory pathways are important components of strategies being developed to induce transplantation tolerance. The aim of this study was to test the effect of a novel chimeric anti-human CD40 monoclonal antibody (Chi 220), either alone or in combination with CTLA4-Ig, on the survival of renal allografts in a nonhuman primate model.

METHODS

Captive-bred adolescent male rhesus monkeys (Macaca mulatta) (4-10 kg) were used as recipients and donors. Four treatment protocols were tested: Chi220 monotherapy, CTLA4-Ig monotherapy, Chi220 combined with CTLA4-Ig, and H106 (anti-CD40L) combined with CTLA4-Ig. Control animals received human albumin. Recipients were followed for survival, renal allograft function as determined by measurement of serum blood urea nitrogen (BUN) and creatinine, chemistries (sodium, potassium, chloride, and bicarbonate), complete blood cell count (CBC) with differential, and the development of donor-specific alloantibody.

RESULTS

Treatment with Chi220 for 14 days prolonged renal allograft survival (MST 38.5 vs. 7 days in untreated controls). Notably, simultaneous blockade of the CD28/B7 pathway did not further augment graft survival but did suppress the development of donor-specific antibodies, an effect not achieved with Chi220 alone, despite peripheral B cell depletion. Finally, treatment with Chi220 suppressed the primary immune response to cytomegalovirus, resulting in severe systemic manifestations.

CONCLUSIONS

Blockade of the CD40 pathway with anti-CD40 mAb is immunosuppressive in a large animal, preclinical renal transplant model. The potential effect of this therapy on viral immune responses will be important to consider for the design of safe clinical trials.

High incidence of cytomegalovirus infection after nonmyeloablative stem cell transplantation: potential role of Campath-1H in delaying immune reconstitution

Nonmyeloablative conditioning is increasingly used for transplantation in a wide range of diseases, but little is known about its impact on the incidence of infections and immune reconstitution. We examined the pattern and outcome of cytomegalovirus (CMV) infections monitored by polymerase chain reaction-based assays and treated preemptively in 101 patients following nonmyeloablative conditioning containing in vivo Campath-1H. Fifty-one patients (50%) had a CMV infection at a median of 27 days after transplantation with a probability of 84.8% in patients at risk of CMV infection. The probability of recurrence of CMV infection before and after 100 days was 53.6% and 46.6%, respectively, and was more common in unrelated donor transplant recipients. All 3 patients who developed CMV disease died of this complication. The 2 patients with late CMV disease had grade III to IV graft-versus-host-disease (GVHD), which occurred de novo in only 4% of patients and in another 10% following donor lymphocyte infusions. The median time to CD4(+) T-cell count more than 200/microL was 9 months in the 48 patients studied. The probabilities of overall survival and nonrelapse mortality at 18 months were 65% and 27.8%, respectively, with no significant difference in survival between CMV-infected and -uninfected patients. The use of Campath-1H appeared to be associated with a low incidence of GVHD but a high incidence of CMV infections and prolonged immune paresis.

Increased risk of complicated CMV infection with the use of mycophenolate mofetil in allogeneic stem cell transplantation

Mycophenolate mofetil (MMF) is increasingly used for prophylaxis and therapy of GVHD in allogeneic stem cell transplantation. In some recent reports of use of MMF in solid organ transplantation a high incidence of CMV disease has been described. We evaluated the frequency and course of active CMV infection in patients who received MMF compared to those who did not receive MMF after allogeneic stem cell transplantation. We retrospectively analyzed 48 adult patients who consecutively underwent unmanipulated allogeneic bone marrow (n = 15) or peripheral stem cell transplantation (n = 33) from HLA-compatible family donors (n = 30) or unrelated donors (n = 18) from February 1997 to September 2000 at our institution. Only patients who were evaluable for the first 100 days were included in this analysis. Sixteen patients received MMF post transplant (MMF+). CMV-antigenemia was monitored by CMV-pp65 antigen. CMV-antigenemia occurred in 14 patients and was virtually only observed in CMV-IgG+ recipients (13/23, 56%). CMV-IgG+/MMF+ patients developed a higher incidence of CMV-antigenemia (8/9, 89%) compared to the CMV-IgG+/MMF- patients (5/14, 35%; P < 0.05). Moreover, five of six patients with persistent or recurrent CMV-antigenemia received MMF. No patient in either group developed CMV disease or died of CMV-related complications. In multivariate analysis including MMF treatment, unrelated vs related donor, GVHD, CMV-serostatus of the donor and stem cell graft type, only MMF treatment was found to be a significant risk factor for both overall and complicated CMV infection.

No cytomegalovirus-related deaths after non-ablative stem cell allografts

Cytomegalovirus (CMV)-related deaths and data of clinically evident CMV disease were assessed in a group of 47 individuals given allogeneic non-myeloablative hematopoietic stem cell transplants (NST). IgG anti-CMV antibodies were found in 56% of the donors and 76% of the receptors. Prophylactic ganciclovir was given to only 12 of the recipients during 100 days after the graft. There were no CMV-related deaths and clinically overt CMV disease was not found in any individual. The follow up post-transplant period of the patients, ranges between 30 and 810 days (median 242 days), the actuarial median survival (SV) is above 810 days and has not been reached, whereas the 810-days SV is 60%. Eighteen patients (38%) died 30-480 days after the transplant; four failed to engraft and died because of progressive disease; three died as a consequence of graft versus host disease (GVHD), whereas eleven individuals had a relapse of the malignancy and died. It is possible that the reduced bone marrow damage during NST, the prompt recovery of both the hematopoiesis and immune function in this type of allografts and the use of peripheral blood hematopoietic stem cell (HSC) is responsible for the absence of CMV-related deaths and clinical disease despite a high prevalence of CMV infection in these individuals.

Unusual viral infections (progressive multifocal leukoencephalopathy and cytomegalovirus disease) after high-dose chemotherapy with autologous blood stem cell rescue and peritransplantation rituximab

Efforts to reduce relapse of non-Hodgkin lymphoma after autologous transplantation have included ex vivo stem cell selection and/or peritransplantation immunotherapy. The late infectious and immunologic consequences of these maneuvers are not well understood, although an increase in early cytomegaloviral disease after CD34(+) stem cell selection and an alteration in immunoglobulin and T-cell recovery after peritransplantation rituximab has been noted. We report the first 2 cases of progressive multifocal leukoencephalopathy caused by JC papovavirus after autologous peripheral blood stem cell transplantation and a case each of cytomegalovirus retinitis and pneumonitis. All 4 patients experienced significant impairment of CD4 T-cell recovery, placing them at risk for these unusual viral infections. The clustering of cases is concerning because all occurred shortly after the introduction of peritransplantation rituximab into treatment protocols (4 of 62 immunotherapy recipients compared with 0 of 276 without; z = 3.595; P <.001), although a direct association with this CD20 B-cell-directed therapy remains speculative.

Orthotopic liver transplantation using low-dose tacrolimus and sirolimus

Although sirolimus (SRL) binds the immunophilin FK506-binding protein-12 (FKBP-12) with greater avidity than tacrolimus (TAC), animal studies have shown that SRL and TAC act synergistically to prevent rejection. Dose-related toxicity is more often the cause of TAC discontinuation than rejection. We hypothesized that SRL would allow for a substantial reduction in the concomitant dose of TAC after liver transplantation to levels less than the threshold for toxicity. A series of 56 liver transplant recipients were administered a combination of SRL and TAC (target trough levels, 7 and 5 ng/mL, respectively). Planned weaning of steroids commenced after 3 months. Pharmacokinetic (PK) studies were undertaken. Patient and graft survival were 52 patients (93%) and 51 grafts (91%), with a follow-up of 23 months (range, 6 to 35 months). One episode (1.8%) of hepatic artery thrombosis was seen. The rate of acute cellular rejection was 14%. No extra treatment was administered in 3 of 8 patients, and the other 5 episodes responded to a single course of steroids. Cytomegalovirus infection occurred in 4 patients (7%). Renal function, glucose control, and lipid metabolism are near normal in 47 patients (84%) without additional medication. Steroid elimination is completed in 51 patients (91%). Bioavailability of SRL and TAC varied between transplant recipients, but trough levels strongly correlated with the area under the curve (r(2) = 0.82 and r(2) = 0.84, respectively). Simultaneous administration did not affect the PK profile of the drugs at this dose. The ratio of trough level to daily dose correlated between SRL and TAC. The synergistic effect seen in animal models also occurs in clinical liver transplant recipients on SRL-TAC combination immunosuppression. A low-dose combination of SRL and TAC should be compared with conventional immunosuppression in a multicenter, randomized, controlled trial.

Valacyclovir for the prevention of cytomegalovirus infection after allogeneic stem cell transplantation: a single institution retrospective cohort analysis

A retrospective single center study was performed to evaluate the safety and efficacy of valacyclovir for prevention of cytomegalovirus (CMV) infection (reactivation) after allogeneic stem cell transplantation (SCT). We compared a group of 31 patients at risk for CMV reactivation (donor, recipient or both seropositive for CMV) who received valacyclovir at an oral dose of 1 g three times a day for CMV prophylaxis with a matched cohort of 31 patients who did not receive the drug or any other form of CMV prophylaxis. Valacyclovir was used as primary prophylaxis in 12 patients and as secondary prophylaxis (after a prior CMV reactivation was effectively treated with either ganciclovir or foscarnet and without CMV antigenemia at the start of valacyclovir) in the remaining 19 patients. The two treatment groups were well matched for the donor-recipient CMV serological status and other pre-transplant characteristics. CMV reactivation was detected by blood antigenemia testing using a commercially available immunofluorescence assay for CMV lower matrix protein pp65 in circulating leukocytes. For primary prophylaxis, 3/12 patients who received valacyclovir reactivated CMV compared to 24/31 patients in the control group (P < 0.001). For secondary prophylaxis, 5/19 valacyclovir patients reactivated compared to 16/24 control patients (P < 0.05). Valacyclovir was well tolerated except for infrequent and mild gastrointestinal side-effects. There was no difference in the incidence of CMV disease in the two groups. Prophylaxis with valacyclovir appears to be safe and efficacious in preventing both primary and secondary CMV reactivation in at-risk patients after allogeneic SCT. Larger prospective randomized studies will be required to confirm these observations.

Fatal upper and lower gastrointestinal cytomegalovirus disease following autologous peripheral blood stem cell transplantation

Although the life-threatening cytomegalovirus (CMV) disease is a well known complication following allogeneic hematopoietic stem cell transplantation (HSCT), it has been considered infrequent after autologous peripheral blood stem cell transplantation (PBSCT). On the other hand, the massive involvement of the gastrointestinal (GI) tract as the primary site of fatal CMV disease is particularly rare after autologous PBSCT. We present the case of a woman who suffered from CMV disease after high-dose busulphan/melphalan/thiotepa (BuMelTT) and autologous PBSCT. The primary site of infection was the GI tract, which was extensively affected. During the fifth week post-transplant the patient started with epigastralgia, diarrhea, fever, GI bleeding, and thrombocytopenia, and she died on day +52. Another case of fatal CMV disease among the few patients treated with BuMelTT has been recently reported, which suggests that the immunodeficiency associated with that regimen can be as intense as that occurring after allogeneic BMT.

Cytomegalovirus Infection and Abdominal Pain with Mycophenolate Mofetil

Mycophenolate mofetil (MMF) is an immunosuppressive agent that exerts relatively selective antiproliferative effects on T and B lymphocytes. Efficacy has been demonstrated in large-scale randomised studies, but the use of MMF is complicated by gastrointestinal upset and is associated with an increased incidence of tissue-invasive cytomegalovirus (CMV) disease. The gastrointestinal tract is a well recognised site for invasive CMV disease, and it has therefore been hypothesised that the abdominal pain commonly seen with MMF is related to CMV infection. This has only been tested in a single small uncontrolled study, where abdominal pain was associated with the presence of CMV on endoscopic biopsy. In contrast, the toxicity profile in 85 patients with psoriasis who had received relatively high dosages of mycophenolic acid, the active moiety of MMF, for up to 13 years showed that the incidence of gastrointestinal upset fell dramatically over time. We can find little evidence that CMV disease explains the gastrointestinal adverse event profile associated with MMF, and instead support the contention that high local concentrations of MMF have a direct toxic effect on cells of the small intestine. We do not recommend any changes to current policy on CMV prophylaxis in patients receiving MMF, although we recognise that some severe gastrointestinal adverse effects may be CMV-associated. The use of trough plasma concentration monitoring, divided doses and a gradually increasing dosage schedule may be of value in limiting toxicity.

Cytomegalovirus colitis during mycophenolate mofetil therapy for Wegener's granulomatosis

Cytomegalovirus (CMV) infection of the gastrointestinal tract is an increasingly recognized cause of morbidity and mortality during the course of HIV infection and in association with immunosuppressive pharmacotherapy. Mycophenolate mofetil, a novel immunosuppressive drug, is currently used in renal transplant recipients and is under evaluation for a variety of disorders. There is preliminary evidence to suggest that CMV reactivation may be more common during treatment with mycophenolate than with other immunosuppressive drugs. We present the case of a 59-year-old male with Wegener's granulomatosis who received mycophenolate and presented with guaiac-positive diarrhea 8 weeks after recovery from Salmonella brandenburg infection. CMV serology and assays for CMV antigens were entirely negative. Colonoscopy demonstrated pancolitis and examination of the specimens disclosed CMV infection. Ganciclovir was administered and the patient made an uneventful recovery. We discuss aspects of gastrointestinal CMV infection with an emphasis on pitfalls in diagnosis and the association with mycophenolate mofetil treatment. We also speculate as to the potential role of previous Salmonella infection and proinflammatory cytokines in CMV reactivation. In summary, when using mycophenolate, clinicians should be more aware of CMV reactivation and disease.

The influence of mycophenolate mofetil on the incidence and severity of primary cytomegalovirus infections and disease after renal transplantation

BACKGROUND

Mycophenolate mofetil (MMF) is increasingly used for immunosuppression after renal transplantation (RTx). The aim of our study was to investigate if the use of MMF has resulted in an increase in the frequency and severity of primary cytomegalovirus (CMV) infections.

METHODS

Retrospective study of adult RTx patients who were CMV seronegative and who received a kidney of a CMV seropositive donor in the period 1992-1997 (n=84). Twenty-four of these patients were treated with MMF (in combination with cyclosporin and prednisone; MMF+) and the other 60 were the control group (cyclosporin and prednisone; MMF-). No CMV prophylaxis was given. CMV infection was defined as CMV seroconversion of IgG antibodies. CMV disease was defined as CMV infection and fever in combination with one or more of the following: leukocytopenia, thrombocytopenia, elevated alanine aminotransferase, or histological evidence of tissue invasive disease.

RESULTS

The incidence of primary CMV infections was similar in both groups (MMF+, 75%; MMF-, 63%). CMV disease was more frequent in the MMF+ group than in the MMF- group (67 vs 30%, P<0.05). In the patients with CMV disease, the use of MMF did not affect severity of symptoms, frequency of tissue invasive disease, or frequency or duration of treatment with ganciclovir.

CONCLUSIONS

Addition of MMF to the immunosuppressive therapy after RTx did not result in an increase of primary CMV infections.However, these CMV infections led more often to CMV disease in patients treated with MMF than in those without MMF.

[Myasthenia gravis. Opportunistic cytomegalovirus infection after long-term azathioprine therapy]

Opportunistic infections after long-term treatment with azathioprine (AZA) have not been noted in patients with myasthenia gravis (MG). We report on a 56-year-old woman with generalized MG who presented with cytomegalovirus infection after being treated with AZA for 17 years. The indication for immunosuppressive treatment in MG should be regularly reconfirmed, particularly since at least 50% of patients can discontinue AZA after two to four years without risk of exacerbation.

Cytomegalovirus infection as a complication of OKT3 therapy in kidney transplant recipients

We compared the incidence of clinical CMV illness in 25 renal transplant recipients treated with OKT3 for steroid resistant cellular rejection with 88 renal transplant patients treated only with conventional immunosuppression (cyclosporin A and steroids). Nine (36%) patients in the OKT3 group developed CMV illness compared to (2.3%) amongst those treated conventionally (p < 0.0005). Patients who received OKT3 were divided into four groups according to the CMV antibody status of the donor and recipient. Six of the 9 episodes of CMV infection occurred in patients not previously exposed to CMV, who received a kidney from a CMV positive donor. Three (12%) of the patients treated with OKT3 died of CMV disease. A further 2 patients died of other causes giving an overall mortality in the OKT3 treated group of 20%. We concluded that when OKT3 therapy is used in association with donor/recipient CMV mismatch it is associated with a high CMV morbidity and mortality.

Immunotoxic effects of hexachlorobenzene on the pathogenesis of systemic, pneumonic and hepatic virus infections in the mouse

A quantitative histopathological method has been developed for the evaluation of the effects of hexachlorobenzene (HCB) on the pathogenesis of three virus infections in the mouse. Hexachlorobenzene was selected because a substantial amount of immunotoxicological data already exists with which we could compare our results. To establish the validity of the method a systemic virus infection (mouse cytomegalovirus, MCMV), a pneumonia causing virus (pneumonia virus of mice, PVM) and a hepatitis virus (mouse hepatitis virus, MHV) were used. We have compared the existing data with the actual pathological effects of hexachlorobenzene on virus disease processes, to gain a more realistic idea of the value of the risk assessment to be derived from extrapolating the in-vitro data in particular, to the in-vivo situation. The results show that the data derived from previous studies on the immunotoxicity of HCB were accurate in predicting the exacerbation of the viral hepatitis, especially in immunodeficient athymic 'nude' mice. It is proposed that this histopathological technique could be a useful technique in the evaluation of host resistance changes following exposure to potentially immunotoxic compounds, but caution will have to be exercised in interpretation in relation to human disease.

Therapy of a fatal murine cytomegalovirus infection with thymic humoral factor (THF-gamma 2) treated immune spleen cells

Infection of mice with murine cytomegalovirus (CMV) presents a model for the study of the role of the immune system in the pathogenesis of human CMV. We performed adoptive transfer experiments to evaluate the prospects for enhancing the anti-viral potential of murine CMV immune spleen cells by THF-gamma 2. Adult BALB/c mice resistant to murine CMV become highly susceptible following immunosuppression by cyclophosphamide. Recipient mice were injected with murine CMV and cyclophosphamide concomitantly, and 24 h later adoptive transfers of syngeneic immune spleen cells were performed. We showed that passive transfers of murine CMV immune spleen cells prevented the development of a fatal disease in 38% of the recipient mice. Daily injections of murine CMV immune donor mice with THF-gamma 2 enhanced considerably (93%) the therapeutic potential of virus-specific immune cells. These experiments provide direct evidence for the antiviral capacity of THF-gamma 2 through its immunomodulatory effect on immune T cells.

Altered susceptibility to viral respiratory infection during short-term exposure to nitrogen dioxide

The studies reported here focus on the relation of nitrogen dioxide exposure to susceptibility to viral respiratory infection in a murine model of pneumonia, created by intratracheal inoculation of an endogenous murine pathogen, mouse cytomegalovirus. The purpose of this work is to clarify the potential role of nitrogen dioxide exposure in the pathogenesis of viral infection of the lower respiratory tract. Previous human epidemiologic studies have presented conflicting information about the relationship of nitrogen dioxide to acute, self-limited episodes of respiratory illness, which are characteristic of viral respiratory infection. Some studies have found an association between exposure to elevated ambient levels of nitrogen dioxide and increased occurrence of acute respiratory illness. In one study this association was found to be strongest in children in the first two years of life. However, other epidemiologic studies have failed to observe this relation. To determine if there is scientific evidence for the possible relation of nitrogen dioxide exposure to human respiratory infection, our studies were performed to assess the impact of nitrogen dioxide on respiratory tract susceptibility to initial, or primary, infection, as well as to recurrent infection, or reinfection, with the identical virus. The latter mechanism of viral respiratory infection is of particular interest, since reinfection is a common method for the development of infection of the lower respiratory tract during early childhood. Outbred CD-1 mice were exposed to either air or nitrogen dioxide for six hours a day on two consecutive days prior to inoculation with murine cytomegalovirus, and then were reexposed to the same level of nitrogen dioxide for six hours a day on four consecutive days, beginning the day after viral inoculation. Susceptibility to primary infection was determined by inoculating animals with an amount of virus (10(2) plaque-forming units) that is too small to produce viral infection in the lungs of normal animals. Mice exposed to 5 parts per million (ppm) nitrogen dioxide routinely developed viral replication in the lung and histologic evidence of pneumonitis after inoculation with this amount of virus, whereas air-exposed animals did not. Most importantly, animals exposed to 5 ppm nitrogen dioxide could be infected with a viral inoculum that was 100-fold smaller than that required to consistently produce viral infection in air-exposed mice. Enhanced susceptibility to infection was found after exposure to 5 ppm nitrogen dioxide, but was not observed with exposure to 2.5 or 1 ppm nitrogen dioxide.(ABSTRACT TRUNCATED AT 400 WORDS)