M. Durand et al., PREDICTION OF RESPIRATORY COMPLICATIONS F OLLOWING ABDOMINAL AORTIC-SURGERY, Canadian journal of anaesthesia, 42(12), 1995, pp. 1101-1107
The most frequent type of complication in patients undergoing aortic s
urgery is respiratory. Preoperative lung function (PFT) and arterial b
lood gas measurement (ABG) are often carried out to assess the risk mo
re precisely. The aim of the present retrospective study was to determ
ine which value of lung function test could identify patients who deve
loped such complications. ''Receiver Operating Characteristic'' (ROC)
curves and the area beneath the curve for the diagnosis of respiratory
complications were calculated for each variable of PFT and ABG. The g
reatest Youden index for each variable was chosen as indicative pulmon
ary function criterion of increased risk of pulmonary complications. O
ne hundred and ninety-five patients (age: 65 +/- 10 years) were includ
ed. Respiratory complications occurred in 15% of patients. Respiratory
complications increased from 12% if the vital capacity (VC) was great
er than or equal to 77% of the predicted value to 35% if the VC was <7
7% (P = 0.002), and from 10% if the FEV(1) was >76% to 34% if the FEV(
1) was greater than or equal to 76% (P = 0.0005). A decreased PaO2 or
increased PaCO2 was not correlated with an increased incidence of resp
iratory complications. Length of stay in ICU or in hospital were incre
ased when VC or FEV(1) were low. Frequency of pulmonary complications
was 9% in patients without PFT abnormalities, 16% in patients with eit
her diminished VC or FEV(1) and 35% in patients with both lowered VC a
nd FEV(1). However, all the areas under the ROC curves were <0.7 and t
he sensitivity of the different variables was low. It is concluded tha
t routine preoperative PFT and ABG cannot predict respiratory complica
tions after abdominal aortic surgery.