Hypovolaemia: non-blood loss: sunken eyes, dry axillae and mucous membranes made it more likely.

Clinical bottom line (level 2a)

  1. Patients with the following were more likely to have hypovolaemia not due to blood loss:
    • sunken eyes (LR+3.4)
    • dry axillae (LR+2.8)
    • dry nose or mouth mucous membranes (LR+2.0)
    • longitudinal furrows on the tongue (LR+2.0)
  2. Patients without the following findings were less likely to have hypovolaemia not due to blood loss
    • moist mucous membranes (LR-0.3)
    • no longitudinal tongue furrows (LR-0.3)
  3. Postural vital sign changes or neurological features were not clearly associated with hypovolaemia not due to blood loss.
McGee et al: Journal of the American Medical Association 1999; 281 (11): 1022-1029
Expires October 2003

The study

Systematic review of all articles of
  • Patients: suspected hypovolaemia
  • Intervention: clinical features
  • Outcome: hypovolaemia
Articles found in English using Medline, 1966 to 1997 (search terms: 3 search strategies based on dehydration, hypotension, orthostatic, heart rate, tilt-table test ) and reviewing bibliographies of retrieved articles and textbooks on physical diagnosis.

Selection criteria: see above
Appraisal criteria: graded using set criteria by 2 independent reviewers
Articles excluded if:
  • studies on children or infants


4 studies of patients with possible dehydration.
Studies were not combined in this analysis.

The evidence


diagnostic test hypovolaemia no hypovolaemia LR+
(95% CI)		 LR-
(95% CI)
sunken eyes (level 1b) - - 3.4
(1.0 to 12.2) 		0.5
(0.3 to 0.7)
dry axilla (level 1b) - - 2.8
(1.4 to 5.4) 		0.6
(0.4 to 1.0)
mucous membranes of mouth and nose dry (level 1b) - - 2.0
(1.0 to 4.0) 		0.3
(0.1 to 0.6)
longitudinal furrows on tongue (level 1b) - - 2.0
(1.0 to 4.0) 		0.3
(0.1 to 0.6)
tongue dry (level 1b-) - - 2.1
(0.8 to 5.8) 		0.6
(0.3 to 1.0)
confusion present (level 1b-) - - 2.1
(0.8 to 5.7) 		0.6
(0.4 to 1.0)
speech not clear or expressive (level 1b-) - - 3.1
(0.9 to 11.1) 		0.5
(0.4 to 0.8)
upper or lower limb weakness (level 1b-) - - 2.3
(0.6 to 8.6) 		0.7
(0.5 to 1.0)
postural hypotension (fall in systolic blood pressure > 20 mmHg) (level 4-) - - 1.5
(0.5 to 4.6) 		0.9
(0.6 to 1.3)
pulse increment > 30 beats/min (level 4-) - - 1.7
(0.7 to 4.0) 		0.8
(0.5 to 1.3)
total

  • Neither mild or moderate postural dizziness nor use of cardiovascular or psychotropic medication was associated with postural hypotension.
  • Poor skin turgor was not found to be clearly associated with hypovolaemia.

Comments

  1. By limiting the search strategy to only Medline and English language articles, potentially useful material may have been missed.
  2. there is no a single gold standard for diagnosing hypovolaemia at bedside (even using invasive techniques*), so misclassification of the true status is a real problem in these studies
  3. The current best option is to go on using the full clinical profile (clinical history, physical and complementary findings) to evaluate these patients, despite this integrative approach having not been quantitatively analysed yet.

Citation

  1. McGee S, Abernethy WB, Simel DL: is this patient hypovolemic?. Journal of the American Medical Association 1999; 281 (11): 1022-1029
Search Terms: hand-search
Contributor: Chris Ball and Clare Wotton, October 1999
Reviewer: Luis Ruiz Del Fresno

Clinical Question.
Patient suspected hypovolaemia or acute blood loss
Intervention or Exposure clinical features; capillary refill, mucous membranes, postural hypotension, pulse
Outcome hypovolaemia or acute blood loss