The ASOS Surgical Risk Calculator: development and validation of a tool for identifying African surgical patients at risk of severe postoperative complications

BACKGROUND

The African Surgical Outcomes Study (ASOS) showed that surgical patients in Africa have a mortality twice the global average. Existing risk assessment tools are not valid for use in this population because the pattern of risk for poor outcomes differs from high-income countries. The objective of this study was to derive and validate a simple, preoperative risk stratification tool to identify African surgical patients at risk for in-hospital postoperative mortality and severe complications.

METHODS

ASOS was a 7-day prospective cohort study of adult patients undergoing surgery in Africa. The ASOS Surgical Risk Calculator was constructed with a multivariable logistic regression model for the outcome of in-hospital mortality and severe postoperative complications. The following preoperative risk factors were entered into the model; age, sex, smoking status, ASA physical status, preoperative chronic comorbid conditions, indication for surgery, urgency, severity, and type of surgery.

RESULTS

The model was derived from 8799 patients from 168 African hospitals. The composite outcome of severe postoperative complications and death occurred in 423/8799 (4.8%) patients. The ASOS Surgical Risk Calculator includes the following risk factors: age, ASA physical status, indication for surgery, urgency, severity, and type of surgery. The model showed good discrimination with an area under the receiver operating characteristic curve of 0.805 and good calibration with c-statistic corrected for optimism of 0.784.

CONCLUSIONS

This simple preoperative risk calculator could be used to identify high-risk surgical patients in African hospitals and facilitate increased postoperative surveillance.

CLINICAL TRIAL REGISTRATION

NCT03044899.

The proteome pattern cGvHD_MS14 allows early and accurate prediction of chronic GvHD after allogeneic stem cell transplantation

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) may be curative, but is associated with significant morbidity and mortality. Chronic graft-versus-host disease (cGvHD), characterized by inflammation and fibrosis of multiple target organs, considerably contributes to the morbidity and mortality even years after allo-HSCT. Diagnosis of cGvHD is based on clinical features and histology of biopsies. Here, we report the generation of a urinary cGvHD-specific proteome-pattern (cGvHD_MS14) established by capillary electrophoresis-mass spectrometry to predict onset and severity of cGvHD as an unbiased laboratory test. cGvHD_MS14 was evaluated on samples from 412 patients collected prospectively in four transplant centers. Sensitivity and specificity was 84 and 76% by cGvHD_MS14 classification. Sensitivity further increased to 93% by combination of cGvHD_MS14 with relevant clinical variables to a logistic regression model. cGvHD was predicted up to 55 days prior to clinical diagnosis. Acute GvHD is not recognized by cGvHD_MS14. cGvHD_MS14 consists of 14 differentially excreted peptides, six of those have been sequenced to date and are fragments from thymosin β-4, eukaryotic translation initiation factor 4γ2, fibrinogen β-chain or collagens. In conclusion, the cGvHD_MS14-pattern allows early, highly sensitive and specific prediction of cGvHD as an independent diagnostic criterion of clinical diagnosis potentially allowing early therapeutic intervention.