The proportion of critical or fatal cases is quite high among hospitalized COVID–19 patients8,35. Although the mortality rate is only 1.4–2.3% based on large-scale epidemiological studies5, about one in three to four hospitalized patients were admitted to ICU4,8,35,36, and 71% to 97.3% of the critically ill patients eventually needed respiratory support8,35–37, while 15% of the ICU patients required extracorporeal membrane oxygenation (ECMO)4. Despite many practices, including respiratory support, different medication regimens, and even ECMO, the case-fatality rate for these critically ill patients was still very high4,8,35–37. Retrospective studies have suggested that the onset of dyspnea was relatively late (median 6.5 days after symptoms onset), but the progression to ARDS could be swift thereafter (median 2.5 days after onset of dyspnea) among patients who developed critical illness36–38. In addition, the high mortality rate in Wuhan in the early outbreak, and in some other areas around the world, exceeded the capacity of local medical resources. These suggest that it is critical to promptly identify patients who are likely to have poor prognosis and higher risk of becoming critically ill.
Although COVID–19 is a multifaceted disease with uncertainty surrounding effective treatments and wide variation in clinical course and prognosis, multiple laboratory features, including lymphopenia, LDH, inflammatory markers, D-dimer, PT, troponin, and CPK, are associated with poor prognosis27,28,39. Our study demonstrates that risk of death among patients with COVID–19 is predictable using a risk score computed from only three common clinical blood samples: LDH, hs-CRP and Lymphocyte (%). As shown in Supplementary Figure S5, these three predictors provided a good separation between survived and deceased patients, in blood samples taken within 10 of patients’ outcome. Front-line clinicians can monitor the disease progression of a patient by applying the proposed risk score to available blood samples. This provides monitoring and screening out high risk patients in real time and as laboratory data become available. Overall, the model serves an accurate indicator for early detection and intervention to reduce the mortality rate, and can potentially monitor the progression of the disease to effectively review and adjust clinical management by healthcare providers.
The significance of our work is five-fold. First, the model may identify high- risk patients early enough and provide them with alternative therapies such as using appropriate respiratory support and other treatments as soon as possible. Second, the model is not based on thresholds and instead provides a continuous probability of death. Thresholds are useful on extreme values, but can be misleading when risk scores are near the thresholds. Instead, probabilities of outcomes provide a level of confidence in the prediction. Third, it provides a simple formula to precisely and quickly quantify the risk of death from just three features of a blood sample. Fourth, the three key features can be conveniently collected at any hospital, even in areas where healthcare resources are limited. The features are objective and quantitative, and therefore avoid any bias of subjective clinical judgements. Last but not least, our research has been constructed using Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD)40 guidance with internal and external validation datasets from multiple centers, and the validation of our model has been confirmed by two cohorts of patients from different hospitals. There are, however, several limitations of the model. First, the patients in the development cohort were from Tongji Hospital, and most were severe and critical. Hence, the cohort may not accurately represent patients with asymptomatic or mild or moderate cases of COVID–19 and the samples could have selection bias. Second, we did not model the effects of different therapies since treatments were not controlled and varied from patient to patient. Finally, this study provided evidence that the risk score could help clinicians determine early intervention for patients with COVID–19 in three Chinese hopitals. We require further investigation and validation involving other hospitals and countries. In particular, it is possible that different hospitals have distinct laboratory, therapies and discharging protocols and that these may affect blood samples and, as a consequence, the interpretation of the risk score.
In conclusion, a simple prognostic risk score system was developed based on a logistic regression classifier to predict death risk for COVID–19 patients, and was validated wth independent cohorts from multiple centers. This risk score system may help healthcare providers to promptly identify patients with poor prognosis and initiate appropriate intervention early to improve the prognosis.