|Home | About | Journals | Submit | Contact Us | Français|
Acute interstitial pneumonia is characterized clinically by ventilation—perfusion mismatching, reduced lung compliance and pulmonary hypertension. Treatment with vasodilators can buy time while the underlying disease is addressed, but these agents have drawbacks as well as benefits.
A man of 50 sought advice after seven days of breathlessness at rest and a non-productive cough. He had stopped smoking 20 years previously. There had been no exposure to industrial dusts or asbestos and he kept no pets. On examination, he was tachypnoeic and centrally cyanosed with digital clubbing. Widespread fine inspiratory crackles were audible throughout the chest. On the chest radiograph there was diffuse ground-glass haziness, whereas six months earlier the appearances had been normal. Total white cell count was raised but there was no eosinophilia. C-reactive protein, serum urea and serum creatinine were also above normal. Antinuclear antigen, antineutrophil cytoplasmic antibody, and avian and fungal precipitants were not detected. Bronchoalveolar lavage yielded necrotic bronchial epithelium but no eosinophils or malignant cells. A CT scan of the chest (Figure 1) confirmed widespread ground-glass opacification with areas of consolidation suggestive of an interstitial lung disorder. Initial therapy with intravenous broad-spectrum antibiotics, bronchodilators and supplemental oxygen led to temporary symptomatic relief. When breathing 60% oxygen the patient was hypoxic with PO2 8.2 kPa, PCO2 5.19 kPa, pH 7.45, HCO3 27 mmol/L, base excess + 3.7.
Continued deterioration necessitated mechanical ventilation. A three-day course of high-dose pulsed methylpred-nisolone was given without effect. Pulmonary artery catheterization revealed pulmonary hypertension (78/35 mmHg) and gas exchange remained deranged despite 100% oxygen (PO2 7.95 kPa, PCO2 9.03 kPa, HCO3 23.4 mmol/L, base excess 6.9, pH 7.185). These findings pointed to a ventilation—perfusion mismatch and hypoxic vasoconstriction of the pulmonary vasculature. In view of this, inhaled nitric oxide (NO) was begun at a concentration of 5-30 parts per million. Gas exchange improved initially (PO2 13.29, PCO2 6.69 kPa, HCO3 31.4 mmol/L, base excess 5.4, pH 7.41, on 100% oxygen) and the pulmonary artery pressure fell to 36/19 mmHg. However, the patient continued to require 100% oxygen so further vascular dilatation was attempted with intravenous prostacyclin, administered at 5 mg/kg. This resulted in systemic desaturation (97% to 91%) and a fall in mean systemic blood pressure (90 to 60 mmHg) and was therefore stopped. NO was continued for two days and oxygen requirement fell gradually to 75%. Oxyhaemoglobin saturation subsequently declined from 94% (on 75% oxygen) to 83% (on 100% oxygen) over 15 hours. Aerosolized prostacyclin was added at 5 ng/kg/min but had no obvious effect on either oxygenation or haemodynamic status. The patient died 10 hours later, and necropsy revealed extensive endstage lung fibrosis.
As in the adult respiratory distress syndrome (ARDS), from which it is clinically indistinguishable, treatment of acute interstitial pneumonia is essentially supportive, aimed at maintaining adequate oxygenation until the inflammatory process abates. In ARDS the use of nitric oxide and prostacyclin2 has been widely reported, so their application in acute interstitial pneumonia is logical. However, the choice and delivery of vasodilator is critical because of potentially serious side-effects. In pulmonary fibrotic conditions there can be large intrapulmonary ventilation—perfusion (V/Q) mismatches.3 Poor aeration of parts of the lung leads to pulmonary vasconstriction and physiological shunting of blood away from these hypoxic areas; however, within these lung regions there are areas that can still participate in gas exchange—hence a V/Q mismatch. Indiscriminate vasodilatation can improve V/Q mismatching but at the same time lead to perfusion of underventilated lung. Treatment with inhaled nitric oxide and prostacyclin avoids this hazard since only ventilated parts of the lung receive the drugs and are thus vasodilated. In our patient, inhaled NO worked initially, with improvement in oxygenation and pulmonary hypertension. As further hypoxia developed, we considered using other vasodilatory agents. Inhaled prostacyclin was thought unlikely to add to the effects of NO since these two agents work on the same underlying pathophysiological mechanism. Intravenous prostacyclin acts as a non-selective vasodilator. When used on its own it results in systemic vasodilatation as well as perfusion of underventilated lung, leading to increased V/Q mismatching and hypoxia4. However, when combined with inhaled NO a synergistic and therapeutic effect has been reported5. Unfortunately, in the case reported here only the non-selective features of this drug combination were evident with systemic hypotension and oxygen desaturation. It is possible that the doses were not ideal. The addition of inhaled prostacyclin was a final attempt to maintain oxygenation, in the hope that it might add slightly to the effect of inhaled nitric oxide. This was unsuccessful either because maximum selective vasodilatation had already been achieved or, more likely, because the disease had entered a terminal phase.