Researchers comparing scoring systems find considerable variation in performance and several factors that complicate direct comparison of systems.
In “Scoring Systems for Outcome Prediction in Patients with Perforated Peptic Ulcer,” published in the Scandinavian Journal of Trauma, Resuscitation, and Emergency Medicine,” the authors reviewed the available scoring systems used for outcome prediction in perforated peptic ulcer (PPU) patients, and evaluated whether “any scoring system has advantages and predictive power to be preferred in clinical practice on this group of patients.”
Because PPU carries a high risk of morbidity and mortality, it is important that clinicians have access to an accurate prognostic scoring system that allows them to properly assess patients and devise an appropriate management strategy, as “variations in the clinical presentation as well as delay in diagnosis and work-up at admission to the hospital, may potentially cause a worsening of symptoms and a deterioration of the clinical condition, with a detrimental outcome.”
For the study, the authors performed a search of the PubMed database to identify studies on PPU patients and related outcomes. They found several studies that used various methods of analysis (receiver operating characteristics curve, area under the curve, etc) to compare a dozen scoring systems that can be used to predict outcomes in PPU patients:
Only the Boey, Jabalpur, Hacettepe, and PULP scores were developed specifically for predicting outcomes in PPU patients; the others have been applied to PPU patients with varying degrees of success.
The authors reported that several factors (including time and sociodemographic differences and differences in score design and complexity) make “direct comparison of results and pooling of patient populations for assessment of outcome prediction difficult or impossible.”
Each of the PPU-specific scoring systems has caveats that should be considered when using the scores in the clinical setting. The Boey score may not be as suitable as others for older patients (age 60 years or older). It also “has not performed well in predicting morbidity.” The Jabalpur and Hacettepe scores were tested in Turkey and India in cohorts that were overwhelmingly male (94% or more) and much younger than typical PPU patients in developed countries. The PULP score “incorporates both the ASA score and the Boey score” and performed slightly better than either of those scores in testing, but still needs to be validated.
Several non-PPU-specific scoring systems were found to have some benefit as predictive tools. The simple-to-calculate ASA score has been “widely used in PPU studies evaluating outcomes” and has been shown to predict mortality well. The MPI score “has been shown to predict both morbidity and mortality, although to a varying degree.” The systems developed initially for use in patients in the ICU (APACHE II, SAPS II, MPM II, POSSUM) are complex, which may limit their implementation in a general clinical setting, but “have all been shown to predict outcome for PPU patients, but to varying degree.”
The authors concluded that “While no scoring system was ideal and all were hampered by certain limitations, a few scores appeared easily applicable in clinical practice. The Boey score and the ASA score are most commonly applied in the current literature to predict outcomes for PPU patients, but both demonstrate variable accuracy. While the PULP score seems promising, a validation is pending before a general application can be recommended.”