Stephan et al developed a clinical score for predicting hypovolemia in a critical care patient. This uses a Spiegelhalter-Knill-Jones weighting for multiple parameters that can help overcome some of the difficulties in estimating the circulating blood volume in this patient population. The authors are from Hopital Tenon in Paris, France.
Parameters (page 755):
(1) pulmonary edema
(2) peripheral edema
(3) skin mottling
(4) congestive heart failure
(5) third spacing (ascites, pleural effusion)
(6) fluid losses from chest tubes, abdominal drainage tubes or nasogastric tubes
(7) fluid balance based on fluid intake and output over the past 24 hours (positive if intake > 400 mL greater than output)
(8) central venous pressure
|
Parameter |
Finding |
Points |
|
pulmonary edema |
yes |
-90 |
|
|
no |
20 |
|
peripheral edema |
yes |
-29 |
|
|
no |
25 |
|
skin mottling |
yes |
29 |
|
|
no |
-10 |
|
congestive heart failure |
yes |
-105 |
|
|
no |
11 |
|
third spacing |
yes |
-184 |
|
|
no |
27 |
|
fluid losses |
yes |
14 |
|
|
no |
-4 |
|
fluid balance |
positive (> 400 mL) |
-24 |
|
|
nil (neutral) |
-24 |
|
|
negative |
41 |
|
central venous pressure |
<= 2 mm Hg |
117 |
|
|
> 2 mm Hg |
-42 |
after Table 4, page 758
where:
• I set a fluid balance of nil in the implementation as 0 – 400 for (intake) – (output). Because of insensitive fluid loss, the lower limit probably should be above 0.
• If data is missing, a zero may be used (page 756). I did not incorporate this into the implementation.
total score =
= SUM(points for all 8 parameters) – 5
probability of hypovolemia =
= 1 / (1 + EXP((-1) * (total score) / 100))
Specialty: Critical Care, Emergency Medicine
ICD-10: ,
