TREATMENT OF BOLIVIAN HEMORRHAGIC FEVER WITH INTRAVENOUS RIBAVIRIN

Download Bolivian hemorrhagic fever (BHF) is a potentially severe febrile illness caused by Machupo virus. (family Arenaviridae). Initial symptoms i...
0 downloads 373 Views 607KB Size
Download PDF
Love PNG Images
718

INTERNATIONAL REPORTS

Treatment of Bolivian Hemorrhagic Fever with Intravenous Ribavirin Paul E. Kilgore, Thomas G. Ksiazek, Pierre E. Rollin, James N. Mills, Mario R. Villagra, Mario J. Montenegro, Maria A. Costales, Luis C. Paredes, and C. J. Peters

From the Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases and the Epidemiology Program Office, Centers for Disease Control and Prevention, Atlanta, Georgia, USA; the National Health Secretary for Bolivia, the German Busch Hospital, and the Japanese Maternal-Infant Hospital, Trinidad, Bolivia; and Viedma Hospital, Cochabamba, Bolivia

Bolivian hemorrhagic fever (BHF) is a potentially severe febrile illness caused by Machupo virus (family Arenaviridae). Initial symptoms include headache, fever, arthralgia, and myalgia. In the later stages of this illness, patients may develop hemorrhagic manifestations including subconjunctival hemorrhage, epistaxis, hematemesis, melena, and hematuria, as well as neurological signs including tremor, seizures, and coma. During the BHF epidemics of the 1960s, convalescent-phase immune plasma from survivors of BHF was administered to selected patients infected with Machupo virus. However, there is currently a paucity of survivors of BHF who can donate immune plasma, and there is no active program for collection and storage of BHF immune plasma; therefore, we had the opportunity to offer intravenous ribavirin to two of three patients with this potentially lifethreatening infection. One patient with laboratory-confirmed Machupo virus infection who received ribavirin recovered without sequelae, as did a second patient with suspected BHF whose epidemiological and clinical features were similar to those of the first patient. This report describes the first use of intravenous ribavirin therapy for BHF in humans, and the results suggest the need for more extensive clinical studies to assess the usefulness of ribavirin for treating BHF.

Bolivian hemorrhagic fever (BHF) was first described in 1959, and there were multiple outbreaks of this infection in the communities of northern Bolivia throughout the 1960s [1]. The etiologic agent is Machupo virus (family Arenaviridae), whose reservoir is the sigmodontine rodent Calomys callosus [2]. Human infections are believed to occur after exposure to Machupo virus in aerosolized secretions or excretions from infected rodents. In 1971, a nosocomial outbreak of BHF that involved five persons occurred in Cochabamba, Bolivia; the investigation of this outbreak suggested that the virus may sometimes be transmitted person-to-person by exposure to fomites, droplets, or aerosols from infected patients [3]. Since the 1960s, the treatment of BHF has consisted of supportive care or, in a small number of cases, the administra-

Received 12 March 1996; revised 16 September 1996. This work was presented in part at the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy held on 17-20 September 1995 in San Francisco. Informed consent was obtained from patients, and the guidelines for human experimentation of the U.S. Department of Health and Human Services and/ or those of the authors' institutions were followed in the conduct of this study. Financial support: This work was supported in part by the U.S. Agency for International Development, La Paz, Bolivia. Reprints or correspondence: Dr. C. J. Peters, Mailstop A-26, National Center for Infectious Diseases, Centers for Disease Control and Prevention, 1600 Clifton Road, N.E., Atlanta, Georgia 30333. Clinical Infectious Diseases 1997; 24:718-22

CD 1997 by The University of Chicago. All rights reserved. 1058-4838/97/2404-0024$02.00

Downloaded from https://academic.oup.com/cid/article-abstract/24/4/718/440018 by guest on 17 July 2018

tion of convalescent-phase immune plasma obtained from survivors of BHF [4]. Previous studies have shown that ribavirin (ICN Pharmaceuticals, Costa Mesa, CA) is effective in the treatment of Lassa fever in humans and other arenavirus infections in animal models; however, ribavirin has not been studied for the treatment of Machupo virus infection in humans [5, 6]. Patients infected with Junin virus, a closely related arenavirus that causes Argentine hemorrhagic fever (AHF), have received intravenous ribavirin in limited trials; some of these patients recovered clinically after receiving this therapy [7]. In September 1994 and October 1994, three patients living in northern Bolivia were identified as having signs and symptoms consistent with BHF. A Centers for Disease Control and Prevention (CDC) medical team that was in Bolivia during September 1994 was dispatched to evaluate the use of ribavirin therapy in patients 1 and 2, while Bolivian physicians evaluated and treated patient 3 in October 1994. The results suggest that more-extensive trials may be warranted to assess the role of ribavirin in the treatment of BHF. Case Reports Patient 1. On 28 August, a previously healthy 34-year-old butcher from Magdalena, Beni Department, Bolivia, developed pyrexia, rigors, and hip arthralgia, followed by widespread myalgia, headaches, and a fever (temperature, 39.2°C). On presentation to the local hospital, his blood pressure was 110/88 mm Hg, and his heart rate was 78; physical examination

CID 1997;24 (April)

Ribavirin for Bolivian Hemorrhagic Fever

showed conjunctival injection and pharyngeal mucosal congestion. Laboratory studies documented a declining total WBC count (count on 29 August, 10,900/mm 3 ; on 30 August, 6,000/mm 3 ; and on 31 August, 4,000/mm 3 ). Four days after the onset of his symptoms, physical examination revealed a diffusely inflamed pharynx with isolated whitish plaques. Because of deterioration in his condition, he was admitted to a tertiary care center in Cochabamba on 3 September for further evaluation and treatment. On admission to the tertiary care center, laboratory investigations revealed a hemoglobin level of 14.5 g/dL, a hematocrit of 44%, and a total WBC count of 4,200/mm 3 . Urinalysis revealed the following values: specific gravity, 1.03; pH, 5; albumin, 2+; hemoglobin, 2+; 30 WBCs per high-power field; and 20-30 RBCs per high-power field. On the third day after admission, the patient developed gross hematemesis and fecal incontinence, followed by generalized seizures requiring treatment with intravenous diazepam. The next day he remained unresponsive and incontinent, and he passed grossly bloody stools. He died later that day after developing oliguria and persistent upper gastrointestinal bleeding followed by shock. The diagnosis of BHF was not considered until the latter portion of his hospitalization. BHF was confirmed after the patient's death by detection of antigens with an ELISA and by isolation of virus from serum specimens. Patient 2. A 52-year-old agricultural worker from San RamOn, Mamore Province, Beni Department, was well until 2 September, when he noted the onset of epistaxis accompanied by fatigue, melena, and hematuria. He was seen by a physician in Trinidad, Bolivia, who admitted him to the hospital on 11 September for evaluation of progressive fatigue and inability to ambulate without assistance. On 12 September, regional public health authorities notified the visiting CDC medical team, who traveled to examine this patient with suspected BHF. Physical examination performed on 13 September while the patient was supine revealed generalized malaise and hypotension (blood pressure, 80/48 mm Hg and heart rate, 72). The mucosa was hyperemic with evidence of recent epistaxis, and both sclera appeared severely injected. The neurological examination showed involuntary left-forearm flexion to 60° as well as a mild resting tremor of the hands, with evidence of global hyperreflexia. Examination of the patellar tendon showed 3+ reactivity, with bilateral spreading to the quadriceps. A finger-nose test showed a gross intention tremor and past-pointing. Initial laboratory studies showed a total WBC count of 1,200/mm 3 , and analysis of a urine sample revealed 2+ protein and 3+ hemoglobin (table 1). On 13 September, the patient's neurological status continued to deteriorate despite supportive care, and ribavirin therapy was initiated 12 days after the appearance of the first symptoms. On the second day of therapy, he developed a cough and expectorated yellowish-brown sputum. Radiographic equipment was unavailable, and he was treated presumptively for pneumonia with amoxicillin. On 21 September he developed a fever (temperature to 39°C) and chills coincident with the presence of hematuria from 18 September to 22 September. Downloaded from https://academic.oup.com/cid/article-abstract/24/4/718/440018 by guest on 17 July 2018

719

Although laboratory facilities for performing cultures were unavailable, we believed that a nosocomial infection of the urinary tract or bronchopulmonary tract represented the most likely source of the pyrexia. The remainder of the patient's hospital course was marked by prolonged hematuria and slow resolution of the neurological signs, even after an 8-day course of therapy. At the conclusion of ribavirin therapy on 23 September, the hematuria had markedly cleared, and he was afebrile and ambulatory. The diagnosis of BHF was confirmed by detection of serum viral antigen and isolation of virus from his serum. Patient 3. A 57-year-old agricultural worker who lived and worked with patient 2 in San Ram& was well until 25 September, when he developed generalized malaise, an intense headache, anorexia, and dysphagia. He presented to a physician in Trinidad on 7 October, and laboratory studies performed that day showed a depressed total WBC count (1,450/mm 3 ) (table 2). From 7 October to 10 October, his condition worsened with the onset of photophobia, diffuse arthralgia and myalgia, fever, and hand tremors. On admission to the hospital on 10 October, physical examination revealed fever (temperature, 39.2°C) and hypotension (heart rate, 76, and blood pressure, 80/50 mm Hg). The large joints of the upper and lower extremities were tender to palpation and on movement around the articulations. A neurological examination showed bilateral resting hand tremors, dysarthria, and an unstable, wide-based gait. Given that the epidemiological characteristics of this patient were similar to those of patient 2 and that neurological signs had become progressive, consistent with a diagnosis of BHF, ribavirin therapy was initiated on 10 October. During the first week of ribavirin therapy, the patient reported gradual resolution of the headache, arthralgia, and myalgia, which was accompanied by a transient decrease in the platelet count (nadir, 57,600/mm' on 12 October) and microscopic hematuria. By day 10 of ribavirin therapy, he was afebrile (temperature, 37°C) with a heart rate of 75 and a blood pressure of 90/70 mm Hg, and he had no evidence of swallowing dysfunction or speech impairment. At the end of a 10-day course of intravenous ribavirin therapy, the patient could ambulate without assistance, but examination showed his gait to be slightly wider-based than normal. No clinical complications were noted while he was receiving ribavirin, but anemia was noted on the final day of therapy and during a follow-up visit on 4 November. Clinical specimens could not be made available to the Centers for Disease Control and Prevention for laboratory confirmation of Machupo virus infection. Discussion

Our recent clinical experience in Bolivia reaffirms the fact that BHF is a life-threatening condition that may mimic other endemic illnesses (e.g., dengue fever and malaria) during its initial phases. Two of the cases described in this report repre-

CID 1997;24 (April)

Kilgore et al.

720

Table 1. Laboratory results for patient 2, a patient with Bolivian hemorrhagic fever who was treated with intravenous ribavirin in September

1994.

Date Laboratory study WBC count (/mm 3 ) Hemoglobin level (g/dL) Hematocrit (g/dL) Platelet count (X10 3 /mm 3 ) Blood urea nitrogen level (normal range, 20-45 g/dL) Creatinine level (normal range, 0.8-1.4 mg/dL) Aspartate aminotransferase level (normal range, 5-35 U/L) Alanine aminotransferase level (normal range, 8-40 U/L) Amylase level (normal level, <120 U/L) Bilirubin level (normal level, <1.0 mg/dL)* Urinalysis

14 September

12 September

1.9

0.7

0.9

1.3

1.5

ND

10,500 12 36 65 50

3,000 12 37 60 20

2,500 13 40 51 25

2,700 12 39 41 28

2,500 14 41 ND 38

1,200 14 45 ND ND

21 September

19 September

17 September

15 September

ND

347

234

113

77

148

ND

156

119

111

95

170

ND

ND

171

158

196

140

ND

ND

1.1/0.6

1.0/0.5

ND

1.7/0.7

2+ Protein 3+ Hemoglobin

4+ Protein 2+ Hemoglobin

2+ Protein 2+ Hemoglobin

ND

Trace protein, trace hemoglobin

2+ Protein, 3+ hemoglobin, 12-15 WBCs, >150 RBCs, granular casts, 3+ bacteria

NOTE. ND = not determined. * Total level/direct level.

sent the first experience in the use of intravenous ribavirin to treat suspected BHF in humans. The clinical and laboratory data obtained during the treatment of these two patients suggest that intravenous ribavirin may be active against Machupo virus following natural infection in humans. While the recovery of patients 2 and 3 coincided with the administration of ribavirin,

we cannot definitively conclude that their recoveries resulted solely from the administration of the drug. Patients 2 and 3 shared exposure histories that were consistent with the known epidemiology of BHF [8]. Both patients were employed by a large ranch to work in open fields containing suitable habitats for C. callosus, the rodent host of

Table 2. Laboratory results for patient 3, a patient with Bolivian hemorrhagic fever who was treated with intravenous ribavirin in October

and November 1994.

Date Laboratory study* WBC count (/mm 3 ) Hemoglobin level (g/dL) Hematocrit (g/dL) Platelet count (X 10 3 /mm 3 ) Aspartate aminotransferase level (normal range, 5-35 U/L) Alanine aminotransferase level (normal range, 8-40 U/L) Urinalysis

7 October

10 October

12 October

14 October

17 October

20 October

4 November

1,500 15 48 170 ND

2,900 14 43 102 ND

2,000 14 44 58 ND

2,000 14 44 65 ND

4,000 12 40 78 ND

5,200 9.6 30 103 184

5,700 8.6 27 296 ND

ND

ND

ND

ND

ND

104

ND

ND

ND

2+ Protein, 1+ hemoglobin, 4-6 WBCs, 8-9 RBCs

2+ Protein, trace hemoglobin, 2-4 WBCs, 3-4 RBCs

No protein, no hemoglobin, 1-2 WBCs

ND

No protein, no hemoglobin, 1-2 WBCs

NOTE. ND = not determined. * Levels of blood urea nitrogen, creatinine, amylase, and bilirubin were not determined.

Downloaded from https://academic.oup.com/cid/article-abstract/24/4/718/440018 by guest on 17 July 2018

CID 1997;24 (April)

Ribavirin for Bolivian Hemorrhagic Fever

Machupo virus. In addition, these agricultural workers were provided with room and board in the same dwelling. Although the exact source(s) of infection in cases 2 and 3 is unknown, each patient may have become infected with Machupo virus from direct exposure to C. callosus while working in the fields, through person-to-person transmission, or by exposure to fomites (e.g., shared utensils) within the ranch dwelling. An additional employee (not described in this report), who was identified during the interview of patient 2, worked and lived with patients 2 and 3 at the same ranch and developed a febrile illness with epistaxis and gingival bleeding, consistent with BHF, in early August 1994. This worker was seen as an outpatient by a local physician in Trinidad and was found to have a depressed total WBC count but was lost to follow-up in late August and did not return to work. Ribavirin is a compound with broad-spectrum activity against selected RNA- and DNA-containing viruses [9]. In the rhesus macaque model for BHF, Stephen et al. demonstrated that ribavirin reduced viremia to nearly undetectable levels by day 10 of therapy and prevented death during the acute hemorrhagic phase of the illness; however, the late neurological signs were not prevented [10]. In another model, mortality among guinea pigs infected with Junin virus was not affected by ribavirin therapy, although viral replication was delayed, and the mean time period until death was increased [11]. Rhesus macaques treated with ribavirin at the time of infection with Junin virus were protected from clinical disease [12]. In a limited trial, the mortality among patients with Argentine hemorrhagic fever who were treated with ribavirin was lower than that among controls, and anemia was the only side effect observed [7]. In a double-blind, placebo-controlled trial of intravenous ribavirin for the treatment of hemorrhagic fever with renal syndrome (caused by Hantaan virus in the family Bunyaviridae), ribavirin reduced the risk of entering the oliguric phase as well as the risk of hemorrhage [13]. In that study, intravenous ribavirin therapy was associated with reversible anemia without sequelae; this side effect had also been identified in previous studies [14, 15]. In a more recent study, intravenous ribavirin appeared effective in the treatment of a laboratory-acquired infection with Sabia virus, an arenavirus related to Machupo virus and first isolated in Brazil [16]. To treat our patients, we used the dosing paradigm that has been used successfully to treat Lassa fever and that was shown by Enria and Maiztegui to be safe for the treatment of Argentine hemorrhagic fever [7]. The anemia in patient 3 appeared to be consistent with both the reversible anemia associated with ribavirin therapy and the pathophysiology of BHF, which may involve suppression of hematopoiesis [7, 17]. The early course of therapy for patient 2 was marked by rapid improvement, followed by a plateau phase in which some neurological signs (e.g., finger-nose test findings) transiently worsened. In the final days of the 10-day treatment period, these neurological sequelae resolved completely. The slower resolution of neurological signs in comparison with that for other abnormalities that was observed for patients 2 and 3, as Downloaded from https://academic.oup.com/cid/article-abstract/24/4/718/440018 by guest on 17 July 2018

721

well as the findings in a study by Enria et al., in which patients with Argentine hemorrhagic fever who received a passive antibody infusion evidenced a late neurological syndrome, suggest direct viral invasion of the CNS rather than an encephalopathic process [18]. Our experience with intravenous ribavirin in Bolivia emphasizes the challenge of conducting clinical trials for the treatment of sporadic diseases in developing countries. Because BHF is no longer an epidemic disease in Bolivia, enrollment of a sufficient number of patients in a randomized trial may be impossible, assuming investigators were to complete such a study in a reasonable period. In addition, ethical concerns about the use of placebo in a randomized study design may arise, since BHF may be life-threatening, and the results of both animal and human studies of Argentine hemorrhagic fever (and animal studies of BHF) suggest that ribavirin is active against arenaviruses. In one alternative study design, investigators could compare drug regimens based on different dosing schedules of intravenous ribavirin to optimize the drug delivery, clinical response, and side-effect profile. On the basis of observations in the 1960s that patients infected with Machupo virus recovered following treatment with convalescent-phase immune plasma from donors, a randomized trial in which treatment with convalescent-phase BHF immune plasma is compared to treatment with intravenous ribavirin could be conducted [4]. This alternative would require a program for screening and storing plasma that has been frozen after collection from a small number of patients who have recovered from BHF. Although hospitals and clinics in the disease-endemic region of Bolivia have limited resources, available laboratories may also provide clinical monitoring for both hematologic and biochemical parameters during ribavirin therapy. Our experience also highlights the advantages of establishing a diagnostic facility in Bolivia that is capable of detecting Machupo virus in clinical specimens during the early phase of BHF. The use of techniques that inactivate Machupo virus in potentially infectious specimens and enable clinicians to make more rapid and cost-effective therapeutic decisions would be required in such a laboratory. Establishment of a Bolivian laboratory in which real-time testing could be done would also enhance surveillance for cases of BHF. If clinicians seek cases more actively and Machupo virus infection can be confirmed in a timely fashion, we may learn the true incidence of BHF and ribavirin can be further evaluated to ascertain if it provides a physically stable, renewable, and safe therapy for the disease.

Acknowledgments

This report is dedicated to the memory of Ronald B. MacKenzie, M.D., M.P.H., who provided invaluable advice and historical insight during preparation of this manuscript. The authors thank Silvia Pozo, M.D., and Joel N. Kuritsky, M.D., for technical assis-

722

Kilgore et al.

tance and John O'Connor, M.S., for editorial assistance in the preparation of this manuscript.

References

1. MacKenzie RB. Epidemiology of Machupo virus infection. I. Patterns of human infection, San Joaquin, Bolivia, 1962-1964. Am J Trop Med Hyg 1965; 14:808-13. 2. Webb PA, Justines G, Johnson KM. Infection of wild and laboratory animals with Machupo and Latino viruses. Bull WHO 1975; 52:493-9. 3. Peters CJ, Kuehne RW, Mercado RR, Le Bow RH, Spertzel RO, Webb PA. Hemorrhagic fever in Cochabamba, Bolivia, 1971. Am J Epidemiol 1974; 99:425-33. 4. Stinebaugh BJ, Schloeder FX, Johnson KM, Mackenzie RB, Entwisle G, DeAlba E. Bolivian hemorrhagic fever: a report of four cases. Am J Med 1966; 40:217 -30. 5. McCormick JB, King IJ, Webb PA, et al. Lassa fever: effective therapy with ribavirin. N Engl J Med 1986; 314:20-6. 6. Peters CJ, Jahrling PB, Liu CT, Kenyon RH, McKee KT Jr, Barrera-Oro JG. Experimental studies of arenaviral hemorrhagic fevers. In: Oldstone MB, ed. Current topics in microbiology and immunology. Vol. 134. Arenaviruses: epidemiology and immunotherapy, Heidelberg, Germany: Springer-Verlag, 1987:5-68. 7. Enria DA, Maiztegui JI. Antiviral treatment of Argentine hemorrhagic fever. Antiviral Res 1994; 23:23-31. 8. MacKenzie RB, Beye HK, Valverde L, Garron H. Epidemic hemorrhagic fever in Bolivia. I. A preliminary report on the epidemiologic and clinical findings in a new epidemic area in South America. Am J Trop Med Hyg 1964; 13:620-5.

Downloaded from https://academic.oup.com/cid/article-abstract/24/4/718/440018 by guest on 17 July 2018

CID 1997; 24 (April)

9. Patterson JL, Fernandez-Larsson R. Molecular mechanisms of action of ribavirin. Rev Infect Dis 1990; 12:1139-46. 10. Stephen EL, Jones DE, Peters CJ, Eddy GA, Loizeaux PS, Jahrling PB. Ribavirin treatment of toga-, arena-, and bunyavirus infection in subhuman primates and other laboratory animal species. In: Smith RA, Kirkpatrick W, eds. Ribavirin: a broad spectrum antiviral agent. New York: Academic Press, 1980:169-83. 11. Kenyon RH, Canonico PG, Green DE, Peters CJ. Effect of ribavirin and tributylribavirin on argentine hemorrhagic fever (Junin virus) in guinea pigs. Antimicrob Agents Chemother 1986; 29:521-3. 12. Weissenbacher MC, Calello MA, Colillas OJ, Rondinone SN, Frigerio MM. Argentine hemorrhagic fever: a primate model. Intervirology 1979; 11:363-5. 13. Huggins JW, Hsiang CM, Cosgriff TM, et al. Prospective, double-blind, concurrent, placebo-controlled clinical trial of intravenous ribavirin therapy of hemorrhagic fever with renal syndrome. J Infect Dis 1991; 164: 1119-27. 14. Canonico PG, Kastello MD, Cosgriff TM, et al. Hematological and bone marrow effects of ribavirin in rhesus monkeys. Toxicol Appl Pharmacol 1984; 74:163-72. 15. Canonico PG. A review of efficacy, toxicity and mechanism of antiviral activity. In: Hahn FE, ed. Antibiotics. Vol 6. Modes and mechanisms of microbial growth inhibitors. New York: Springer-Verlag, 1983: 161-86. 16. Barry M, Russi M, Armstrong L, et al. Brief report: treatment of a laboratory-acquired Sabia virus infection. N Engl J Med 1995; 333:294-6. 17. Child PL, MacKenzie RB, Valverde LR, Johnson KM. Bolivian hemorrhagic fever: a pathologic description. Arch Pathol 1967; 83:434-45. 18. Enria D, Franco SG, Ambrosio A, Vallejos D, Levis S, Maiztegui J. Current status of the treatment of Argentine homorrhagic fever. Med Microbiol Immunol 1986; 175:173-6.