© 2003 American Thoracic Society
Acute Respiratory Failure after Interferon-
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
-1b (200 µg subcutaneously) was given on March 15th (Day 1) and was repeated on Day 3 and Day 5. Paracetamol was given for a mild flu-like syndrome. Transient liver cytolysis (alanine amino transferase 288 IU/L, normal less than 50 IU/L; aspartate amino transferase 549 IU/L, normal less than 37 IU/L) occurred on Day 6. IFN and paracetamol were stopped. On Day 8, dyspnea increased, and nasal oxygen (2 L/min) was started. On Day 14, IFN-
was resumed (200 µg on Days 14, 16, and 18) and then stopped. On Day 15, the dyspnea worsened, with mild fever (38°C) without sputum. The complete blood cell count and liver function tests were normal. Blood cultures, urine culture, and Legionella pneumophila serology were negative. Chest radiography and high-resolution CT (Figure 1)
showed diffuse ground-glass opacities superimposed on preexisting reticular opacities. Spiral CT excluded proximal pulmonary embolism. The patient was admitted to the intensive care unit on Day 17. Swan-Ganz catheter gave normal results (pulmonary artery pressure 30/2023 mm Hg, wedged pulmonary artery pressure [PAPw] 10 mm Hg, cardiac index [CI] 2, 58 L/min/m2). Broad-spectrum antibiotics and high-dose intravenous methylprednisolone (80 mg/day on Days 19 and 20, and 1,000 mg/day on Days 21, 22, and 23) were begun. Mechanical ventilation with orotracheal intubation was initiated on Day 28, with a PaO2/FIO2 ratio below 140. Bronchoalveolar lavage was negative for bacteria, mycobacteria, and Pneumocystis carinii. The patient died on Day 49 due to refractory hypoxemia. Necropsy showed diffuse acute alveolar damage lesions with preexisting UIP; no granulomas, abscesses, or viral cytopathic effect was found (Figure 2)
. There was no evidence of pulmonary embolism. No infectious agents were found by direct examination with the usual stains (hematoxylin eosin, periodic acid Schiff [PAS], Grocott and Ziehl).
|
|
-1b was initiated on 6 April, 2001 (200 µg three times per week), together with prednisone (20 mg/day) and paracetamol. Two months later the respiratory condition had worsened. The patient was evaluated for lung transplantation. Pulmonary hemodynamics measured with a Swan-Ganz catheter were normal (PAP 32/18 mm Hg, PAPw 10 mm Hg, CI 3.48 L/min/m2). His condition worsened suddenly on 28 June, after 35 doses of IFN-
, with increased dyspnea, productive cough, and fever (up to 39.5°C). On admission, the patient had diffuse crackles and cyanosis. Blood chemistry showed isolated elevated C-reactive protein (220 mg/L, N less than 6 mg/L). The blood white cell count was normal. Blood and urine culture was negative. Serologic tests for Mycoplasma pneumoniae, Chlamydia pneumoniae, and L. pneumophila were negative. Chest radiography showed increased ground-glass opacities. Arterial blood gas analysis showed severe hypoxemia (PaO2 37 mm Hg, PaCO2 37 mm Hg). IFN was stopped. Combined antibiotherapy and nasal oxygen (5 L/min) were started. The fever resolved within 24 hours, but his respiratory condition did not improve. He was admitted to the intensive care unit on June 30. High-dose intravenous methylprednisolone (500 mg/day for 3 days and then 80 mg/day) and intravenous anticoagulation were begun without improvement. Right-lung transplantation was performed on July 11. Analysis of the explanted right lung showed diffuse alveolar damage with preexisting UIP (Figure 3)
. No signs of pulmonary embolism, granulomas, abscesses, or a viral cytopathic effect were found, and direct examination showed no microorganisms with the usual stains (hematoxylin eosin, PAS, Grocott and Ziehl). The patient died a few weeks later.
|
-1b (200 µg subcutaneously) was initiated on March 14. Within 6 hours after the first injection, he developed fever (39°C) and chills and increased hypoxemia. Nasal oxygen was increased to 8 L/minute. Fever resolved within 12 hours. Blood cultures gave negative results. A second injection of IFN-
-1b (100 µg) was given on March 16. Again, fever occurred within 6 hours, accompanied by severe dyspnea. Chest radiography showed increased alveolar opacities in the lower lobes. Nasal oxygen was increased to 12 L/minute, and IFN was stopped. The ECG, total blood cell count, and blood chemistry were unchanged. Blood and urine culture was negative. Intravenous furosemide, anticoagulation, intravenous methylprednisolone (120 mg/day over 4 days) and broad-spectrum antibiotics were ineffective. L. pneumophila serology and urinary L. pneumophila 1 antigen were negative. He died on March 20, 7 days after the first IFN-
injection, with refractory hypoxemia. The family refused permission for autopsy.
Patient 4
A 69-year-old man, an ex-smoker (5 pack-years) with a personal history of retinitis pigmentosa and familial history of pulmonary fibrosis (one brother and two cousins), received a clinical diagnosis of IPF in July 2001 while living in Argentina. Prednisone (60 mg/day for 3 months, tapered to 25 mg/day) and azathioprine (75 mg/day) were given without improvement. He was seen at Bichat hospital in July 2002 because of worsening dyspnea. Physical examination showed bilateral inspiratory crackles and digital clubbing. High-resolution CT showed bilateral reticular opacities with honeycombing without ground-glass opacities. Lung function tests showed a severe restrictive defect with impaired gas exchanges and resting hypoxemia (Table 2). Standard biologic tests were normal. Echocardiography was normal. Fiberoptic bronchoscopy could not be performed because of poor respiratory tolerance. Azathioprine was stopped. Prednisone was reduced to 20 mg/day, and IFN-
(200 µg subcutaneously) was initiated on July 12, together with paracetamol. Within 12 hours after the first injection he developed fever (39°C), chills, arthralgias, and myalgias, which lasted for 24 hours. Dyspnea increased. On July 15, the second injection of IFN-
(200 µg) was again followed by fever and increasing dyspnea. Profound hypoxemia required high oxygen flows (PaO2 35 mm Hg, PaCO2 29 mm Hg on nasal oxygen 7 L/minute). Chest radiography showed extensive alveolar opacities in the lower lobes. The ECG pattern was unchanged. Mild thrombocytopenia was noted (120 x 106/mm3). Blood and urine culture was negative, as were blood tests for antibodies against M. pneumoniae, C. pneumoniae, and L. pneumophila. IFN was stopped. High-dose intravenous methylprednisolone (120 mg/day for 4 days) was given without improvement. He died on July 23, 11 days after the first IFN-
injection, with intractable hypoxemia. The family refused autopsy.
Comparison of Patients Who Did and Did Not Develop Acute Respiratory Failure After Starting IFN Therapy
Pre-IFN pulmonary function tended to be worse in the patients who developed an acute respiratory failure (Table 2). TLC was below 45% in three of the four patients with acute respiratory failure but only in one of the six control patients. Similarly, diffusing capacity for carbon monoxide was below 30% in three of the four patients with acute respiratory failure and in two of the six control patients. All four patients with acute respiratory failure had either TLC of less than 45% or TLCO of less than 30% before starting IFN-
therapy.
The interval between diagnosis of IPF and initiation of IFN therapy tended to be shorter in the four patients who developed acute respiratory failure (Table 1), and the mean dose of prednisone given with IFN-
tended to be higher (p = 0.06).
DISCUSSION
IFN-
was recently proposed (3) as a treatment for IPF. We observed four patients with advanced pulmonary fibrosis who developed an irreversible acute respiratory failure shortly after initiation of IFN-
.
Definite IPF was diagnosed in patients 1 and 2 according to ATS/ERS criteria, as usual interstitial pneumonia was proven by open lung biopsy (1). Probable IPF was diagnosed in patient 3, who fulfilled all major and minor ATS/ERS criteria but did not have surgical lung biopsy (1). Patient 4 lacked one IPF criterion (bronchoscopy could not be done) but had typical clinical and CT features of IPF. All the patients had worsening lung function despite steroid treatment. Acute respiratory failure occurred in three patients shortly after initiating IFN-
(after two injections in two patients and after six injections in one patient) and in one patient after 35 injections. The clinical and radiologic pattern was very similar in each case, with increasing dyspnea, fever, and rapidly progressive hypoxemia requiring high oxygen flows. Chest radiography (performed in every patient) and thorax CT (one patient) revealed new alveolar opacities. No other cause was found despite further investigations. Empiric treatments, including antibiotics, diuretics, anticoagulation, and high-dose corticosteroids, were ineffective. All four patients died of respiratory failure, after lung transplantation in one case (patient 2). The lung could be analyzed at necropsy in patient 1 and at the time of transplantation in patient 2. In both cases, pathologic analysis showed diffuse alveolar damage with preexisting UIP.
Acute exacerbations of IPF, with a histologic pattern of acute diffuse alveolar damage superimposed on UIP, have previously been described (8, 9), and sometimes respond to high-dose corticosteroids. The mechanism of these acute exacerbations, often considered idiopathic, is unclear, but infection seems an unlikely cause (8). In our opinion, IFN-
-1b therapy was the most likely cause of acute respiratory failure in our four patients. First, respiratory failure occurred in three cases shortly after the initiation of IFN-
administration. Second, no other cause of deterioration was found, even by pathologic examination of the lungs in two cases. Left ventricular failure was excluded by invasive assessment of pulmonary hemodynamics during IFN-
therapy in two cases (patients 1 and 2). Patients 3 and 4 had normal left ventricular function, as assessed by echocardiography before initiating IFN-
, and these patients had no clinical signs of heart failure and no electrocardiographic changes at the onset of acute respiratory failure. Third, IFN-
administration has previously been associated with acute respiratory failure in humans. In a phase I trial of recombinant IFN-
combined with recombinant interleukin-2 in patients with cancer, pulmonary toxicity (with rales and dyspnea), related to the dose of both IFN-
and IL-2, was reported (5). Several studies have shown a higher incidence of radiation pneumonitis in patients treated with IFN-
(4, 6). In one study, the incidence of severe or fatal pneumonitis was 44% in patients treated with IFN-
combined with radiotherapy, compared with 9.5% in similar patients receiving radiotherapy alone. Moreover, in animal models IFN-
exacerbates lung inflammation (10) and is necessary for the expression of hypersensitivity pneumonitis (11). IFN-
has also been shown to induce acute exacerbations of autoimmune diseases in humans (12). Taken together, these data suggest that IFN-
has the potential to trigger the development of an acute alveolar injury in some pathologic settings.
A rapid corticosteroid dose reduction may also have contributed to the onset of respiratory failure in two of our patients. However, in patients 2 and 4, acute respiratory failure occurred even though the corticosteroid dose was maintained at 20 mg/day. Furthermore, the six control patients who had no respiratory adverse effects on IFN-
received lower doses of prednisone than did the patients with acute respiratory failure. In their original article, Ziesche and colleagues reduced the prednisolone dosage from 50 to 7.5 mg/day within 14 days before initiating IFN-
therapy (3). However, the possibility remains that the risk of a rapid corticosteroid dose reduction may be higher in patients with more advanced lung disease.
The exact frequency of IFN-
associated acute respiratory failure is unknown. Ziesche and colleagues described no acute exacerbations of IPF in their preliminary study of nine patients treated with IFN-
(3). Raghu and associates reported in the form of an abstract their experience with IFN-
therapy in 29 patients with IPF (13). Four patients, all with end-stage disease, died within 4 weeks of starting IFN-
therapy, and one patient died after 3.5 months of this treatment (13). The potential link between IFN-
therapy and early death was not discussed in this abstract.
Our four patients with fatal acute respiratory failure appeared to have poorer pre-IFN pulmonary function than the six control patients (Table 2) and the patients studied by Ziesche and colleagues (3). In the latter study, the mean TLC value was 70% in the IFN-
group. In our overall study population, four of the seven patients with either TLC of less than 45% or TLCO of less than 30% developed an acute interstitial pneumonia. The severity of the restrictive pattern could have contributed to the poor tolerance of acute respiratory failure. Alternatively, advanced lung fibrosis may be associated with some biologic particularities that could explain the onset of acute respiratory failure.
Furthermore, two of our patients had a history of familial lung fibrosis. Whether this might interfere with the action of IFN-
is uncertain. However, in view of recent data linking some cases of familial pulmonary fibrosis with abnormal processing of the surfactant protein C precursor protein, the pathophysiology of familial pulmonary fibrosis might differ from that of sporadic IPF (14).
Pending further studies, we recommend that IFN-
therapy be avoided in patients with end-stage pulmonary fibrosis and in patients with familial pulmonary fibrosis.
Acknowledgments
The authors thank Dr. Isabelle Herry for critical review of the manuscript.
Received in original form August 7, 2002; accepted in final form December 24, 2002
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Proc. Am. Thorac. Soc. | Am. J. Respir. Cell Mol. Biol. |