To the editor: The interesting article by van Beers et al. states that functional studies does not offer clues as to the presence of mild pulmonary hypertension (PH) in patients with sickle cell disease (SCD) [1]. However, they did show that although not statistically significant, lung function findings do vary in hemoglobin-SS patients.
Studies examining lung function in adults with SCD conclude that pulmonary function is abnormal in 90% of patients and gradually declines over time [2,3]. The most common patterns observed were restrictive lung disease and low diffusing capacity (D LCO ). Few publications include data on both the D LCO and the transfer coefficient (K CO ). However, measuring K CO provides information on pathophysiology, which cannot be obtained just from the D LCO . These measurements help us understand when lung disease is due to a loss of lung units (e.g., pneumonectomy), a process interfering with the alveolar surface (e.g., fibrosis) or the capillaries (e.g., PH).
Studies on patients with pulmonary arterial hypertension show that a low D LCO can be used as an early marker of PH, before echocardiographic signs develop [4]. When patients with SCD are screened with echocardiogram up to 36% have evidence of PH depending on the hemoglobin genotype [5,6]. However, in autopsy studies there is histological evidence of PH in up to 75% of patients [7]. D LCO may therefore be a more reliable tool than echocardiogram in predicting PH. Van Beers study shows both the D LCO and K CO were reduced in patients with mild PH [1].
We examined lung function results from 32 adults with SCD, average age was 34 years and 30 patients had hemoglobin-SS. The mean forced expiratory volume in 1 sec (FEV 1 ) was 81.8% predicted and the mean forced vital capacity (FVC) was 80.9% predicted. Two patients had an obstructive defect (FEV 1 /FVC < 70%) and 18 patients had a FEV 1 /FVC >80%. The mean D LCO (82.6% AE 3.2%) was lower than predicted but the mean K CO (110.4% AE 3.2%) was higher than predicted. There was no statistical difference between patients with or without a history of acute chest syndrome. The results suggest that in these patients the pathophysiological process may be the loss of lung units.
To our knowledge, only two other studies on adults with SCD included data on K CO [8,9]. These studies had similar patient numbers, characteristics and spirometry results. Unlike our study both these studies found the K CO lower than predicted. This implies in these patients a different pathophysiological mechanism was operating. The difference with our patient group may be one of disease severity. The high K CO we found is consistent with an increased blood flow to normal lung. A lower than predicted K CO could imply the patients had developed PH, as suggested by the findings in van Beers study, or diffuse alveolar damage. Only two studies performed in children examined D LCO and K CO and found results similar to our study, perhaps reflecting their lower exposure to pulmonary complications [10,11].
We feel both our study and the van Beers study supports including K CO and D LCO in routine lung function tests to help us understand the pathophysiology and also predict early damage.