QJM Advance Access originally published online on March 9, 2009
QJM 2009 102(5):335-339; doi:10.1093/qjmed/hcp020
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sitting and standing blood pressure measurements are not accurate for the diagnosis of orthostatic hypotension
From the Clinical Age Assessment Unit, Mid-Western Regional Hospital, Dooradoyle, Limerick, Ireland
Address correspondence to Dr J. Cooke, Clinical Age Assessment Unit, Mid-Western Regional Hospital, Dooradoyle, Limerick, Ireland. email: john.cooke{at}hse.ie
Received 11 November 2008 and in revised form 20 January 2009
 |
Abstract |
|---|
Introduction: Orthostatic hypotension (OH) is associated with troublesome symptoms and increased mortality. It is treatable and deserving of accurate diagnosis. This can be time consuming. The current reference standard for its diagnosis is head-up tilt (HUT) testing with continuous beat-to-beat plethysmography. Our objective was to assess the accuracy of sit-stand testing with semi-automatic sphygmomanometry for the diagnosis of OH.
Design: Retrospective test of diagnostic accuracy.
Methods: This was a retrospective study performed using a database maintained by a busy syncope unit. HUT testing was performed using an automated tilt table with Finometer monitoring. A 3 min 70° HUT was performed following 5 min supine. Sitting blood pressure (BP) was measured following 3 min rest. Standing BP was measured within 30 s of assuming the upright posture. The results of sit-stand testing were compared with HUT testing as a reference standard. Both tests happened within 5 min of each other and patients underwent no intervention between tests.
Results: From a total of 1452 consecutive HUTs, we identified 730 with pre-test measures of sitting and standing BP. The mean age of this group was 70.57 years (SD = 15.1), 62% were female. The sensitivity of sit-stand testing was calculated as 15.5%, specificity as 89.9%, positive predictive value as 61.7%, negative predictive value as 50.2% and the likelihood ratio as 1.6. The area under the Receiver Operator Curve was 0.564.
Conclusion: We have demonstrated that sit-stand testing for OH has very low diagnostic accuracy. We recommend that the more time-consuming reference standard method of diagnosis be used if the condition is suspected.