Impaired pulmonary function in discharged patients with COVID-19: more work ahead.

From the authors:

We are grateful for the opportunity to respond to the letter by S. Nusair, whom we thank for his interest and thoughtful remarks about our study on the abnormal pulmonary function in coronavirus disease 2019 (COVID-19) patients [1].

Diffusing capacity of the lung for carbon monoxide (DLCO) reflects the capacity of carbon monoxide transfer from the environment to the pulmonary capillary blood, which is the most clinically useful routine methodology to evaluate the function of the lung to exchange gas. KCO is the rate constant for carbon monoxide uptake from alveolar gas, which is influenced mostly by the thickness and area of the alveolar capillary membrane, the volume of blood in capillaries supplying ventilated alveoli, and the concentration and properties of haemoglobin in the alveolar capillary blood. DLCO is mainly determined by KCO and the alveolar volume (VA). Mathematically, KCO can be calculated as DLCO/VA under body temperature, ambient pressure, saturated with water vapour conditions [2].

We agree with S. Nusair that it should be noted that DLCO/VA is not a simple ratio and in fact the relationship between lung volume and carbon monoxide uptake is linear and certainly less than 1:1 [3]. Recent studies have tended to use KCO instead of DLCO/VA, as it may be inferred from the use of the term DLCO/VA that DLCO can be normalised for VA. The 2017 European Respiratory Society/American Thoracic Society standards for single-breath carbon monoxide uptake in the lung recommends that the term of KCO is preferred to DLCO/VA [2]. However, in practice, many pulmonary function test reports from different manufacturers still present with DLCO/VA and not KCO. Besides, in order for an easier and direct comparison with some previous studies on pulmonary function in patients with severe acute respiratory syndrome (SARS) [4, 5], DLCO/VA was kept in our report. It is important to understand the interpretation of KCO and DLCO/VA, but the term of KCO is recommended for more popular use in future.

Decreased KCO occurs in alveolar-capillary damage, microvascular pathology or anaemia. Causes of low VA include reduced alveolar expansion, alveolar damage or loss, or inspired gases maldistributed with airflow obstruction. When KCO turns normal, in the presence of a low DLCO it is associated with low VA. Because only the functional alveolar units had been sampled, a biased picture toward more preserved areas of the lungs is thereby provided [3]. Crucially, the same DLCO may occur with various combinations of KCO and VA, each suggesting different pathologies. It is difficult to interpret which is the predominant role because both decreased alveolar volume and KCO contribute to the pathogenesis of impaired diffusion capacity. DLCO gives an overall assessment of gas exchange for the lung, while the alveolar-capillary membrane diffusing capacity for carbon monoxide (DMCO) only dependents on molecular diffusion of the membranes. However, DMCO can be calculated using different equations but is not yet standardised, and as such is not a commonly used gas diffusing parameter [6].

In our study, 51 patients with COVID-19 had impaired DLCO, including 29 patients who showed both impaired DLCO and KCO. The number of patients with both impaired DLCO and KCO (29/51) was 24% larger than the number of patients with impaired KCO alone (22/51). Moreover, the percentage of patients with decreased total lung capacity was less than the percentage of patients with decreased KCO. Furthermore, vascular injury and thrombosis have been demonstrated to be important contributing factors in the pathogenesis of COVID-19. SARS coronavirus 2 infects the endothelial cells via angiotensin-converting enzyme 2. The recruitment of immune cells to the endothelial cells facilitates the induction of endotheliitis in several organs, which contributes to systemic impaired microcirculatory function [7]. Acute pulmonary embolism has been reported during the clinical course [8]. Post mortemstudies have reported vascular injury and thrombi in multiple organs [7, 9]. Ackermann et al. [10] compared the lung pathology of COVID-19 with H1N1. The result showed that patient with COVID-19 presented distinctive vascular features, consisting of severe endothelial injury associated with the disruption of intercellular junctions, cell swelling and a loss of contact with the basal membrane. From the above, it is probable that the effect of membrane lesion plays an important role on the impaired diffusion capacity in patients with COVID-19.

Our study showed that, similar to DLCO, DLCO/VA (representing the transfer factor of carbon monoxide) also showed an association with the disease severity: as the patients progressed to a more severe condition, the mean value of DLCO/VA compared to predicted values was lower. Since the standard deviation values suggest that individual differences have effects on lung function, we also provided the distribution of the DLCO/VA of predicted in patients with different severities. 42% of the subjects with severe pneumonia had a DLCO/VA of less than 80% of predicted, compared to only 27% in the patients with pneumonia and 13% in patients with mild illness, respectively.

In previous follow-up studies of SARS, impaired DLCO was defined as when DLCO was <80% of the predicted value, irrespective of KCO or the alveolar volume. In follow-up studies of rehabilitating SARS patients ranging from 0.5 to 2 years, the impaired DLCO was the most common abnormality, accounting for 15.5% to 43.6% [11-13]. The mean DLCO ranged from 77.7% to 93.8% pred in the studies above, with the longest follow up of 2 years. Our study showed a mean DLCO of 78.2% pred at time of discharge but follow-up data is not available yet. It is unclear how the impaired diffusion capacity changes in survivors of COVID-19, as well as how the changes may be reflected on chest computed tomography (CT). In one study, where surviving SARS patients were seen 6 months after admission, 43(75.4%) of patients showed radiological abnormalities, including 62.8% who had abnormal DLCO [11]. Thus, the lung function impairment findings are consistent with the CT images of SARS survivors showing persistent ground glass opacity, reticular opacities and traction bronchiectasis suggesting the fibrosis [14]. Likewise, in discharged patients with COVID-19, Wang et al. [15] reported that 94% (66/70) of patients who were discharged from hospital at the end of the study still had mild to substantial residual lung abnormalities on their last CT scans.

Our study was a pilot study which first described the impaired pulmonary function in patients with COVID-19 at time of discharge. Impairment of diffusion capacity is the most common abnormality followed by restrictive ventilatory defects. However, more follow-up work is needed to assess the longitudinal variation of these deficits.

This one-page PDF can be shared freely online.

Shareable PDF ERJ-02194-2020.Shareable