Positive predictive value and trends of inferior petrosal sinus sampling (IPSS) in diagnosing cushing disease and ectopic ACTH secretion: A systematic review and meta-analysis

Inferior petrosal sinus sampling (IPSS) offers a means of differentiating between Cushing disease and Cushing syndrome with lower false-positive and false-negative rates relative to traditional techniques. However, consolidated data on efficiency reflecting contemporary use is lacking. We present a comprehensive meta-analysis of IPSS as a means of diagnosing ACTH-cortisol axis derangements via both CRH and desmopressin-stimulated techniques. Searches of 7 electronic databases from inception to December 2020 were conducted following PRISMA guidelines. Articles were screened against pre-specified criteria. Outcomes were pooled by random-effects meta-analyses of proportions where possible. We performed a meta-analysis of sixty-eight unique publications, assessing each technique for positive predictive value (PPV), false positive rates, and overall changes in practice patterns over time. A total of 68 studies satisfied all criteria, with 3685 (3471, 94.2% confirmed) and 332 (285, 85.8% confirmed) patients tested for Cushing’s disease and syndrome, respectively. Pooled analyses demonstrated an overall PPV of 89.3% (95%CI[83.6%, 94.0%]) in CRH stimulation diagnosis of Cushing disease. In desmopressin stimulation, our analyses demonstrated an overall PPV of 96.5% (95%CI[94.5%, 98.1%]) in diagnosis of Cushing disease. There was a significant decline in the use of CRH-stimulation IPSS in diagnosis of both Cushing disease (p = 0.0055) and Cushing syndrome (p = 0.013). Concurrently, there was a significant increase in the use of desmopressin-stimulation IPSS in diagnosis of both pathologies (p < 0.0001). Our findings demonstrate significant changes in practice patterns with respect to IPSS stimulation technique. Our pooled analyses demonstrate improved diagnostic performance in desmopressin stimulation procedures relative to CRH stimulation procedures. Further multi-institutional studies with special attention to acquiring quality data for sensitivity, specificity, and other critical analyses are necessary to truly evaluate this promising technique.