|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Internal Medicine (Drs. Castillo, Naranjo, Sepúlveda, and Ossa), Temuco Teaching Hospital and Faculty of Medicine Universidad de la Frontera, Temuco, Chile; and University of New Mexico Health Sciences Center (Dr. Levy), Albuquerque, NM.
Correspondence to: Constanza Castillo, MD, Department of Internal Medicine, Medicine Faculty, Universidad de la Frontera, M. Montt 116 Temuco, Chile; e-mail: eureka{at}telsur.cl
| Abstract |
|---|
|
|
|---|
Design: A retrospective chart review abstracting clinical, radiologic, laboratory, and epidemiologic data.
Setting: ICU of the university teaching hospital in Temuco, Chile.
Patients: Sixteen patients with HPS treated between 1997 and 1999.
Results: Patients were aged from 19 to 45 years, 82% were men, and 88% were farm or timber workers with occupational acquisition of HPS. After an incubation period ranging from 5 to 25 days, a prodromal influenza-like phase frequently was accompanied by abdominal symptoms. From 1 to 7 days later, respiratory insufficiency and hemodynamic instability suddenly appeared. In 81%, hemorrhage was evident; in 63%, moderate-to-severe bleeding occurred. The most prominent laboratory abnormalities were hemoconcentration, leukocytosis, thrombocytopenia, altered partial thromboplastin time (PTT), creatine kinase, transaminases, and hyponatremia. Creatinine elevation was common, with clinical importance in two patients. All patients had severe hypoxemia and pulmonary edema. Fifteen patients received supportive treatment, and 5 patients were treated with corticosteroids. The mortality rate was 43.8%.
Conclusions: Bad prognostic factors appeared to be severe hypotension, lower PaO2/fraction of inspired oxygen values, prolonged PTT, hemorrhage, greater volume load, and profuse bronchorrhea. The effects of treatment with corticosteroids could not be determined. Hemorrhage and renal involvement were common in our patients, features not often described in the North American literature of Sin Nombre virus HPS.
Key Words: Andes virus Hantavirus pulmonary syndrome hemorrhagic disorders South American Hantavirus
| Introduction |
|---|
|
|
|---|
In 1994, isolated cases of HPS were reported in Brazil, followed by nearly 20 cases in Paraguay. In 1995, an epidemic outbreak began in the southern part of Argentina and Chile.17 It is believed that climate changes, related to "El Niño" oceanic stream, favored the overgrowth of rodents in rural areas, especially in the Andes mountain chains. A new strain of Hantavirus was identified and called Andes virus, with the primary host, the wild rat Oligoryzomys longicaudatus.18 19 20
In Temuco, Chile, 20 serologically confirmed cases (positive IgM antibodies detected by enzyme-linked immunosorbent assay [ELISA] following the Centers for Disease Control and Prevention [CDC] guidelines)21 of HPS have been treated: 16 adults and 4 children. We describe the epidemiologic and clinical features of the 16 adults.
| Materials and Methods |
|---|
|
|
|---|
The clinical charts were reviewed by the authors who also attended the patients and conducted the treatment procedures. Laboratory measurements were done using the same technique in all patients at the Temuco Hospital laboratory. The chest radiographs were interpreted by the attending physician at the ICU, and conclusions were confirmed by a radiology consultant. Epidemiologic data were produced by personnel of the Ministry of Health with expertise in epidemiology of infectious diseases who researched the most probable place and time of the HPS acquisition in each of the HPS cases, using a three-pathway model based on clinical histories, the account of their relationships, and information from coworkers. They also visited the most probable sites of HPS acquisition to verify the contagion in a specific site. The incubation period was estimated based on this information. Two-sample Wilcoxon rank-sum (Mann-Whitney) test or Fishers Exact Test was performed, and p < 0.05 was considered statistically significant.
| Results |
|---|
|
|
|---|
The mean age of the patients was 30 years (range, 19 to 45 years), 82% were men, 88% were Hispanics, and 12% were South American natives. The estimated incubation period was calculated based on the most probable place and time of infection reported by the epidemiologists of the Ministry of Health, and ranged between 5 and 25 days.
The mean period of the prodromal symptoms was 4 days (range, 1 to 7 days). All suffered from fever and myalgia, 50% had GI symptoms, 30% had headache, 30% had dry cough, 25% had back pain, and 13% had dizziness, polyarthralgia, and odynophagia. One patient had diarrhea, another had conjunctival injection, and one had a maculopapular rash. From 1 to 7 days (mean, 4 days) after the initial prodromal symptoms, sudden hypotension (systolic BP < 120 mm Hg) and respiratory distress appeared. During the advanced stage of the illness, the most common findings were tachycardia (> 100 beats/min), tachypnea (> 25 breaths/min), and cough (100%). Hypotension and fever were present in 94%. One patient was hypothermic (temperature < 35°C) at the time of ICU admission. Bleeding disorders were seen in 81% of the patients during the advanced state of the disease (Tables 1 , 2 ). Major external bleeding (> 500 mL) was observed in 63% of the cases. All six fatal cases had clinical evidence of bleeding (Tables 1 , 2) .
|
|
|
Urinalysis was performed in 15 patients. In five patients, a RBC count on urinalysis confirmed the clinical report of hematuria (Table 2) . In three additional samples, the RBC count was high (> 10 RBCs per high-power microscopic field or > 50/mL) indicating microscopic hematuria (Table 2) . None of these eight patients underwent urinary catheterization before the specimen was obtained.
Arterial blood gas analyses were available in 14 patients. Thirteen of them showed hypoxemia and one had hypoxemia with carbon dioxide retention (PaCO2, 72 mm Hg). The mean ratio of PaO2 to the fraction of inspired oxygen (FIO2) was 139 mm Hg at the day of ICU admission (range, 44 to 296 mm Hg; Table 4 ). The patient who died without arterial blood gas measurement had an oxygen saturation < 70% on 100% oxygen. In all cases, the laboratory results showed the most altered values during the first 24 to 48 h of the cardiopulmonary stage and improvement thereafter within 4 to 7 days in the survivors.
|
Treatment
All patients received oxygen between 0.4 and 1.0
FiO2, and 69% received mechanical
ventilation with 7 mL/kg tidal volume and positive end-expiratory
pressure levels ranging from 5 to 12 cm H2O. All
patients received IV fluids. The mean volume administered on the day of
ICU admission was 3.2 L (range, 1 to 10.6 L; Table 4 ). Vasoactive drugs
were administered to 10 hypotensive patients (dobutamine, dopamine,
epinephrine, phenylephrine, and norepinephrine). Monitoring with a
pulmonary artery (Swan-Ganz) catheter was possible in five patients
receiving dopamine or associated vasoactivedrugs and parenteral volume
(four survivors and one nonsurvivor). The pulmonary artery occlusion
pressure (PAOP) was normal at ICU admission in four of these five
patients (4 to 12 mm Hg), and one of them had PAOP of 19 mm Hg after
vigorous volume loading. The PAOP rose at the nadir of the
cardiopulmonary stage in three patients (16 to 19 mm Hg). One survivor
and one nonsurvivor had decreased systemic vascular resistance index
(SVRI): 1,289 dyne·s/cm5·m2 and 1,678
dyne·s/cm5·m2, respectively (normal range,
1,700 to 2,500 dyne·s/cm5·m2) with normal
cardiac index (CI) at ICU admission (normal range, 2.5 to 4.2
L/min/m2). A third patient showed a slightly increased SVRI
(2,729 dyne·s/cm5·m2) with normal CI at ICU
admission. After administration of vasoactive drugs, the SVRI increased
slightly, but the CI dropped in all these three patients at the nadir
of the cardiopulmonary stage (1.63 L/min/m2, 1.94
L/min/m2, and 1.94 L/min/m2, respectively). The
other two patients continued with normal hemodynamic values. In five
patients, who responded rapidly to volume loading and the
administration of dopamine, the insertion of a pulmonary catheter was
avoided. In other five critically ill patients (four of them with
hemorrhages), the insertion of a central catheter was extremely
difficult and not possible in three patients. One patient was admitted
to the ICU in cardiopulmonary arrest. Methylprednisolone was
administered to five patients (1 g the first day, 500 mg the second and
third days, and 250 mg on the fourth and fifth days).
Clinical Course
Seven of the 16 patients died, giving a case fatality rate of
43.8% (Table 4)
. The nonsurvivors had lower systolic BP at the time of
ICU admission and severe hypoxemia; the
PaO2/FIO2
ranged from 44 to 68 mm Hg (mean, 55 mm Hg; Table 4
). The PTT, platelet
count, and other laboratory test results were more severely abnormal in
the nonsurvivor group (Table 4)
. Six of the 7 patients who died had
major external bleeding (Table 2)
. Platelet counts and PTT values did
not predict the appearance of bleeding or the severity of the
hemorrhage, but were less deranged in three patients who had no
bleeding (mean PTT, 78 s vs 97 s; nadir platelet count,
44 x 103/µL vs
32 x 103/µL).
The fluid volume administered in the patients who died was larger than in the survivors (Table 4) . Four of the patients, who did not survive, were in refractory circulatory collapse and probably received excessive IV fluids (> 4 L; Table 4 ). These patients eliminated large amounts of serous fluid through the endotracheal tube at the ICU admission day (up to 3 L in one patient); this bronchial secretion was not seen in the survivors (Table 4) . In 10 hypotensive patients (4 survivors and 6 nonsurvivors), vasoactive drugs were administered. Five of the nine survivors received mechanical ventilation as well as six of seven nonsurvivors (Table 4) . The one patient who did not receive mechanical ventilation and died was admitted in cardiorespiratory arrest and died before he was connected to the ventilator. Five of nine survivors received methylprednisolone (Tables 4 , 5 ).
|
| Discussion |
|---|
|
|
|---|
During the cardiovascular stage, which appeared around the fourth or fifth day after the initial prodromal symptoms, pulmonary edema developed suddenly. On the chest radiographs, pulmonary edema shadows were evident in all patients, but we did not find pleural effusions described as frequently as in the North American patients.23 Patients in whom massive pulmonary edema developed in < 24 h from ICU admission did not survive. The speed of radiologic worsening seems to be a bad prognosis factor, and it was associated with the worsening of PaO2/FIO2 values. Patients who developed refractory shock died in < 24 h despite the use of inotropic vasoactive drugs and volume loading. Hemodynamic measurements in five patients receiving inotropic drugs and parenteral fluids showed normal PAOPs and normal or low CI, as described in the North American literature,6 11 but decreased SVRI in two patients. We did not use extracorporeal membrane oxygenators, a therapeutic approach that seems to be useful, because we did not have these resources.24 25
One of the major findings of our study is that during the cardiovascular stage of the disease, 81% of the patients had hemorrhage. In 63% of them, hemorrhage was moderate to severe, and most of these patients bled from more than one source. In four patients, hemorrhage was extremely severe and preceded death. In the North American literature, microscopic hematuria and prolonged PTT were described, but severe clinically significant hemorrhage is not mentioned.6 Hemorrhages were not described in the Argentinian HPS cases.22 Hyponatremia as a manifestation of the disease was present in 69% of our patients. There was no correlation between hyponatremia and the amount or composition of the administered fluids. Altered renal function test results were seen in 54% of our patients, and two patients showed overt renal failure. No correlation between hyponatremia renal function was found. Renal failure was not described in the first North American HPS patients,6 7 8 9 10 11 but high serum creatinine levels were reported in the Argentinian HPS casuistic22 as well as in some North American HPS patients.6 The Andes virus seems to trigger a mixed clinical picture between the variants isolated in the North American and the Asian Hantaviruses. Linderholm et al26 described pulmonary involvement in patients with hemorrhagic epidemic fever. In 1996, HPS with renal failure caused by the Black Creek Canal virus was also described.13 In the North American patients, hyperamylasemia was seen and attributed to ribavirin.27 Our patient with hyperamylasemia received neither ribavirin nor amantadine. The global mortality was 43.8%. Patients who died suffered severe hemodynamic compromise and catastrophic respiratory failure, which worsened in 12 to 24 h despite the use of mechanical ventilation and vasoactive drugs. The survivors stabilized their hemodynamic failure and respiratory distress in 24 to 48 h (fifth to sixth day of the initial prodromes) and recovered in the following 4 to 7 days.
Five patients received methylprednisolone, all of whom were among the survivors. There is a trend toward lower creatinine and clinical hemorrhage, higher platelet count, but longer PTT in steroid-treated patients, but the small number of patients does not allow us to make conclusions about the utility of steroids in this disease, and further controlled studies are necessary.
Person-to-person transmission of HPS was documented in Argentina.17 However, in the Coihayque, XI region of Chile, the seroprevalence of antibodies against Hantavirus in health-care workers was 3.7%, similar to the general population in that community.20 We studied 67 health-care workers at the Temuco Hospital in the IX region of Chile who attended an average of four critically ill HPS patients, and none of them showed Hantavirus antibodies.28 Despite intrafamilial clusters of HPS in Chile, the person-to-person transmission of Hantavirus has not been proven in our country.
| Footnotes |
|---|
Financed by the Universidad de la Frontera Temuco, Chile. Registration No. 99/110.
Received for publication June 23, 2000. Accepted for publication January 23, 2001.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |