Tactile vocal fremitus (TVF) is the vibration sensation felt by the examining hand during palpation of the patient’s chest wall (specifically the intercostal spaces) while the patient speaks (e.g. saying “toys for tots” or “ninety-nine”). It is commonly performed as part of the chest examination to detect pathologies such as pneumothorax or pleural effusion (decreased TVF) or consolidation (increased TVF). However, similar findings can be elicited by listening with a stethoscope to vocal resonance (VR) during lung examination. In one study using hospitalized patients with respiratory symptoms, TVF yielded likelihood ratio (LR) findings that were similar to VR for detecting pleural effusion (has LR+ of 5.7 versus 6.5 when diminished and LR- of  0.2 versus 0.3 when normal, for TVF versus VR respectively).² Additionally, TVF has been found to be clinically useful only in patients with pleural effusion.  In other words, one could omit TVF during lung exam and instead listen to VR while auscultating lung fields to note for any asymmetry. Listening with the stethoscope is less subjective than palpating with hands and beginners in physical exam training might find VR easier to interpret and more practical than TVF. Also, a busy medical resident probably will prefer not to spend time doing a second examination test to elicit the same finding. Thus, TVF could be abandoned.

The liver span is performed by percussion of the right hemothorax together with palpation of the right upper abdominal quadrant and is used to estimate liver size at the bedside. It is usually taught as part of the examination of the abdomen. However, this technique has been found to have non-significant LRs.³ Also, liver size is often underestimated by this method and physicians vary in their interpretation of their findings.⁴ In most cases when liver pathology is suspected, imaging (at least an ultrasound) of the liver will still be required to better delineate the size and textures of the liver to make a proper differential diagnosis. Frequently, I encounter learners doing or commenting on liver span during standardized testing scenarios but I cannot remember a clinical encounter with real patients where I really wanted the medical student or resident to tell me his/her finding about liver span.  Therefore, Liver span could also be dropped from the list of physical exam skills that need to be taught.

Kernig's sign and Brudzinski's sign are popular among medical students in physical exam evaluation of patients with fever and headache, a common reason for hospital admission worldwide, especially when meningitis is in the differential diagnosis. However, in 190 patients admitted with acute encephalitis syndrome⁵ and 297 patients with suspected meningitis,⁶ these two signs performed poorly in discriminating between patients with or without meningitis as determined by subsequent cerebral spinal fluid analysis, and their LRs were close to 1. So, in clinical practice they may be of little help in clinical decision making and, given the discomfort they can cause for patients, they can probably be dropped from the clinical exam checklist, as well.

Homans sign for deep venous thrombosis (DVT) also seems well-known among medical students who tend to mention it during presentation of cases of pulmonary embolism or DVT. Homans sign, or the dorsiflexion sign, is elicited by forceful dorsiflexion of the leg and noting whether there is increased resistance (or calf pain, although some authors claim that pain was not described in the original paper by Homans).  A positive test was thought to indicate DVT, but clinical utility was found to be doubtful with non-significant LRs.^7,8 Other validated scoring systems that rely on other lower extremity exam findings (e.g. Wells scoring system for DVT) are widely available and more reliable in making the diagnosis. Therefore, and since it may be associated with patient discomfort and theoretical risk of dislodging an existing clot, physicians should discourage learners from attempting to elicit this sign during lower extremity examination of patients suspected to have DVT.

This paper was supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through UCSF-CTSI Grant Number UL1 TR001872. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

The author declares that there are no conflicts of interest.