Differentiating epileptic seizures and pseuodseizures

June 4, 2011

In a study of 120 events from 35 consecutive patients referred for video electroencephalography (VEEG) monitoring:

Video-documented predictors of psychogenic nonepileptic seizures were:

  • “preserved awareness,” “eye flutter,” and “bystanders can intensify or alleviate”)

Video-documented predictors of epileptic seizures were:

  • “abrupt onset,” “eye-opening/widening,” and postictal “confusion/sleep”

However, “eyewitness reports of these 6 signs were inaccurate and not statistically different from guessing.”

Clinical examination for peripheral arterial disease

June 2, 2011

Summary:

If both pulses on both feet are normal and there are no femoral bruits, there is a less than 10% chance of an ABI being abnormal in a study population with a prevalence of disease of 28%. I could not make any other conclusions from this article.

Methods:

Patients:

1619 patients who has undergone peripheral arterial testing between December 2005 and February 2010. Excluded:

  • 228 due to an abnormally high ABI (greater than 1.30) in either leg
  • 155 patients because the examination for femoral bruit was not recorded
  • 1 patient because the examination for distal pulses was not recorded

Reference standad:

  • ABI <= 0.9

Results:

  • 1236 patients (2472 legs)
  • API was abnormal in at least on leg of 348 patients (28%)
  • 575 patients had a normal clinical examination (no femoral bruits and both pulses on both feet normal) and 546 of these (95%) has normal ABIs. I calculate the confidence interval is 92.75 % to 96.54 %.
  • The authors state that a complete clinical examination has a sensitivity and specificity of 58.2% and 98.3%; however, I cannot reproduce these numbers. The problem may be due to the authors inconsistently reporting number of patients versus number of legs and also not clearly stating whether an abnormal examination is when any one finding is abnormal or when all findings are abnormal.

Citation

Armstrong DW, Tobin C, & Matangi MF (2010). The accuracy of the physical examination for the detection of lower extremity peripheral arterial disease. The Canadian journal of cardiology, 26 (10) PMID: 21165366

Carnett’s test for excluding intra-abdominal origins of abdominal tenderness

May 26, 2011

Interpretation of this study is difficult since the authors excluded patients with functional gastrointestinal disorders. I will continue to use Carnett’s sign as one vote among components of the evaluation of abdominal pain.

Methods

Patients and examiners

130 consecutive outpatients with abdominal tenderness. “Patients who had functional
gastrointestinal disorders such as functional dyspepsia or irritable bowel syndrome were excluded from the analysis because of the potential for comorbidity by psychiatric disorders.”

Two of 8 generalist physicians with a mean of 7 years in practice examined each patient. One examiner knew the patient’s medical history and one was blinded.

Description of the test

“After routine examination of the abdomen, the site of maximum
tenderness is determined. The patient is then asked to
contract the abdominal muscles by raising his/her head from
the examination table while the examiner continues to apply
pressure to the tender site. The test is positive if tenderness
becomes more severe or is unchanged. A positive test suggests
that the abdominal wall is the source of pain. On the
other hand, the test is negative when tenderness is reduced,
which suggests that the pain has an intra-abdominal source.”

Reference standard

Final diagnosis after clinical evaluation.

Results

Results are in the Figure. 86% of patients with abdominal wall or psychogenic pain had a positive test. 13% of patients with intra-abdominal causes of pain had a positive test.

The reproducibility was high, κ=0.81.

Citation

Takada T, Ikusaka M, Ohira Y, Noda K, & Tsukamoto T (2011). Diagnostic usefulness of Carnett’s test in psychogenic abdominal pain. Internal medicine (Tokyo, Japan), 50 (3), 213-7 PMID: 21297322

Digiti quinti sign for hemiparesis

February 11, 2011

Patients: Sixty patients with unilateral brain tumors without obvious focal signs and 30 controls

Examination: A physical therapist, blinded to all clinical and  imaging data, performed 13 clinical tests. A photograph of the digiti quinti sign is at http://pubmed.gov/20799429. In this test, the patient extends arms as for the pronator drift test, but turns palms down. In an abnormal test, the 5th finger on the weak side abducts more than the 5th finger on the normal side.

Outcome: MRI

Results:

Selected results from 13 tests that were studied.
Test Sensitivity Specificity
Digiti quinti sign 51% (41-61) 70% (61-79)
Pronator drifting test 41% (31-51) 96% (92-99)
Babinski sign 8%(2-14) 100%

Comment: this study used a case control design and so is subject to spectrum bias. However, restricting all patients to those without obvious focal signs may somewhat compensate.

This has been added to http://en.citizendium.org/wiki/Digiti_quinti_sign

Citation

Maranhão ET, Maranhão-Filho P, Lima MA, & Vincent MB (2010). Can clinical tests detect early signs of monohemispheric brain tumors? Journal of neurologic physical therapy : JNPT, 34 (3), 145-9 PMID: 20799429

Distinguishing categories of shock

February 11, 2011
Patients: 68 adults with more than  30 minutes of systolic blood pressure <90 mmHg
Examination: performed by medicine house officers.
Outcome: Category of shock classified by a single, blinded, experienced clinician using published consensus guidelines when available.
Results: The combination of estimation of the skin temperature (patient’s hand cooler than the examiner’s hand) and central venous pressure (CVP >7 cm) correctly categorized 52 of 68 cases (accuracy = 76%).
Although not explicitly stated, the categorization seems to be:
  • Septic shock (37 patients): warm hands; CVP  < 8 cm.
  • Cardiogenic shock (18 patients): cool hands, CVP  > 7 cm.
  • Hypovolemic shock (13 patients): cool hands, CVP  < 8 cm.

Comment: The authors used an odd definition of pulse pressure, “same/wider vs. thinner than examiner’s”, which may have prevented this sign from being significant. Other studies have found that a proportional pulse pressure < 25% is predictive of poor cardiac physiology (PMID: 2913385; PMID: 11420761)

This has been added to http://en.citizendium.org/wiki/Shock_(physiology).

Citation

Vazquez R, Gheorghe C, Kaufman D, & Manthous CA (2010). Accuracy of bedside physical examination in distinguishing categories of shock: a pilot study. Journal of hospital medicine : an official publication of the Society of Hospital Medicine, 5 (8), 471-4 PMID: 20945471

Cervical neck rotation for headache diagnosis

January 14, 2011

Patients: The authors performed a cross sectional study of 60 patients who came from three groups of 20 patients: migraine headache, cervicogenic headache, and mixed headache.  The examiners were not aware of the patients’ diagnoses.

Examination: The cervical flexion–rotation test (FRT) maneuver “consisted of pre-positioning the cervical spine in maximal end range flexion followed by passive rotation of the head to the left and the right, with the subject relaxed in supine. End of range was determined either by firm resistance encountered by the therapist or the subject reporting the onset of pain, whichever came first.” The examination was done on a symptom-free day.

Results: The best cutoff with 30 degrees of rotation; patients with less than 30 degrees were more likely to have cervicogenic headache. The authors found that the area under the receiver operating characteristic curve was 0.85; however, the authors did not provide the sensitivity and specificity for that cutoff value.

Comment: The results of the study may be inflated. It is not clear whether the authors studied consecutive patients with headache or if the study was subject to spectrum bias. In addition, the patients with cervicogenic headache 10 years older which may influence cervical mobility.

Citation:

Hall TM, Briffa K, Hopper D, & Robinson K (2010). Comparative analysis and diagnostic accuracy of the cervical flexion-rotation test. The Journal of Headache and Pain, 11 (5), 391-7 PMID: 20508964

Diagnosing Gout

September 27, 2010

Clinical Diagnosis of Gout

Patients: In a study of 381 patients recruited by family physicians because of monoarthritis, 57% had positive crystals and one patient had a bacterial infection.

Results: The family physicians had an accuracy of:

  • Sensitivity 97%
  • Specificity 28%

The authors then created a prediction rule for the 328 patients that the family physicians suspected had gout:

Male sex 2.0 points
Previous patient-reported arthritis attack 2.0 points
Onset within 1 d 0.5 points
Joint redness 1.0 points
MIP involvement  2.5 points
Hypertension or and cardiovascular diseases 1.5 points
Serum uric acid level >5.88 m9/dL  3.5 points

The prediction rule had an accuracy of:

  • 8 or more points
    • Sensitivity 92% (193/209)
    • Specificity 88% (77/88)
  • 4 or less points
    • Sensitivity 99% (208/209)
    • Specificity 50% (44/88)

If 4 or less points, the NPV is 98%.

If 8 or more points, the PPV=95%.

Comment: Even if using the prediction rule developed in this study, physicians should consider aspirating most all joints with monoarthritis as value even in joints suggestive of gout had a one in 20 chance of an alternative diagnosis. Also, the label of gout may affect future decision making and life-long medications.

References:

1. Janssens HJ, Fransen J, van de Lisdonk EH, van Riel PL, van Weel C, & Janssen M (2010). A diagnostic rule for acute gouty arthritis in primary care without joint fluid analysis. Archives of Internal Medicine, 170 (13), 1120-6 PMID: 20625017

2. Peláez-Ballestas I, Hernández Cuevas C, Burgos-Vargas R, Hernández Roque L, Terán L, Espinoza J et al. (2010) Diagnosis of chronic gout: evaluating the american college of rheumatology proposal, European league against rheumatism recommendations, and clinical judgment. J Rheumatol 37 (8):1743-8. DOI: 10.3899/jrheum.091385 PMID: 20551101.

Diagnostic accuracy of confrontation visual field tests

April 28, 2010
Diagnostic accuracy of confrontation visual field tests
among 332 eyes with a prevalence of defects of 57%
Selected findings Sensitivity (%) Specificity (%) Kappa
Red comparison
77
27
0.74
Kinetic 5 mm red target
74
93
0.67
Finger comparison
71
57
0.69
Static finger wiggle
44
97
0.63
Finger counting
25
100
0.79

How to do the tests

  • Red comparison: “Two identical red atropine bottle tops, approximately 20 mm in diameter, were presented in a fashion analogous to the finger comparison test described above and the patient was asked if the bottle tops appeared equally red. Any quadrant in which the bottle top appeared less red was considered abnormal.”
  • Kinetic red target: “A 5-mm red-topped pin was moved inward from beyond the boundary of each quadrant along a line bisecting the horizontal and vertical meridians. The patient was asked to report when the pin was first perceived to be red.”
  • Finger comparison: “The examiner’s index fingers were presented simultaneously on either side of the vertical meridian in the superior and then inferior quadrants approximately 20° eccentric to fixation and the patient was asked to report if the fingers appeared equally clear. Any quadrant in which the finger appeared less clear was recorded as abnormal.”
  • Static finger wiggle: “Two index fingers were presented simultaneously on either side of the vertical meridian approximately 20° eccentric to fixation and equidistant from the quadrant borders in the superior and then inferior quadrants. The patient was asked to report which finger wiggled (<5° oscillation).”
  • Finger counting: “The patient was asked to count 1 or 2 static fingers presented sequentially in each of the 4 quadrants approximately 20° eccentric to fixation and equidistant from the quadrant borders.”

Reference

Kerr NM, Chew SS, Eady EK, Gamble GD, & Danesh-Meyer HV (2010). Diagnostic accuracy of confrontation visual field tests. Neurology, 74 (15), 1184-90 PMID: 20385890

Using the physical examination to predict response to fluid bolus.

March 18, 2010

This post, and the last post to ClinDx about heart failure, use the physical examination not as a traditional diagnostic test, but rather to help predict response to treatment.

The current study shows that an increase of pulse pressure of ≥ 9% (as measured by a radial artery catheter) can predict an increase in stroke volume after 500 cc of  6% hydroxyethyl starch among patients with severe sepsis or pancreatitis.  The differences in the areas under the ROC curves were no significantly different among the three findings in the table.

Predicting increase in cardiac stroke volume after intravenous infusion of 500 cm of fluids
Finding after passive leg raising Sensitivity Specificity
Increase ≥ 9% in pulse pressure as measured by radial arterial catheter 79% 85%
Increase ≥ 8% in femoral artery blood flow as measured by Doppler ultrasonography 86% 80%
Increase ≥ 10% in stroke volume as measured by bedside echocardiography 86% 90%

A limitation is that the pulse pressure was measured by radial artery catheter. Presumably, an auscultated blood pressure would perform similarly, but this is not certain.

This has been added to http://en.citizendium.org/wiki/Head-down_tilt.

Using the physical exam to direct chronic treatment of heart failure

March 3, 2010

Targeting a clinical score to a score of 2 or less based on the Framingham diagnosis of heart failure with the following findings may reduce mortality similarly to targeting a NT-proBNP level and better than usual care:

  • Major findings (1 point each)
  • Minor findings (0.5 point each)
    • Orthopnea
    • Reduction in exercise tolerance
    • Resting heart rate > 100 bpm
    • Hepatomegaly
    • Peripheral edema

This has been added to http://en.citizendium.org/wiki/Heart_failure#Goals.

Citation:
Lainchbury JG, Troughton RW, Strangman KM, Frampton CM, Pilbrow A, Yandle TG, Hamid AK, Nicholls MG, & Richards AM (2009). N-terminal pro-B-type natriuretic peptide-guided treatment for chronic heart failure: results from the BATTLESCARRED (NT-proBNP-Assisted Treatment To Lessen Serial Cardiac Readmissions and Death) trial. Journal of the American College of Cardiology, 55 (1), 53-60 PMID: 20117364