Diagnostic accuracy of clinical tests for the detection of splenomegaly

Abstract

Background: Due to common pain referral patterns, splenomegaly may present similarly to other conditions commonly seen in the physical therapy setting, and may become deadly if undetected. Therefore, when considering a differential diagnosis for left upper quadrant pain and the appropriateness of physical therapy intervention, it is imperative to screen for splenomegaly.Objective: To summarize the research on the diagnostic accuracy of clinical examination tests for the detection of splenomegaly to determine if physical therapists and other health care providers can rely upon these tests to screen for splenomegaly.Methods: A computer-based search strategy was conducted, using PubMed (MESH terms), Cochrane Central Register of Controlled Trials, ProQuest Nursing & Allied Health Source, SPORTDiscus, CINAHL, and SCOPUS databases (key words). Selection required: (1) an assessment of individual clinical measures of splenomegaly, (2) diagnostic accuracy measures for clinical tests, (3) an acceptable reference standard for comparison, and (4) a case–control or case-based case–control design. Electronic and hand search limits included English-written and human studies.Results: The studies included Nixon’s percussion, Castell’s percussion, percussion of Traube’s space, bimanual palpation of the spleen, ballottement of the spleen, Middleton’s maneuver, and the combination of percussion and palpation testing. For each clinical test from each study, diagnostic accuracy measures were reported as well as reliability, when available. Quality scores were tabulated for each study using the QUADAS instrument and studies were dichotomized into high (≧10/14) or low (<10/14) quality based on QUADAS scores. Thirteen articles qualified for inclusion, of which four were identified as high quality. When comparing high quality studies, palpation techniques yielded specificity values of 82–93 and sensitivity values of 31–85·6, whereas percussion techniques presented with lower values in both specificity (32·6–87) and sensitivity (22–85·7).Conclusion: There is moderate evidence for the use of palpation tests for ruling in splenomegaly and limited evidence to support the use of palpation or percussion techniques for ruling out this condition. More high quality studies are needed.