Secondary Logo

Journal Logo

Original Article

Respiratory effects of the kneeling prone position for low back surgery

Radstrom, M.; Loswick, A. C.; Bengtsson, J. P.

Author Information
European Journal of Anaesthesiology: April 2004 - Volume 21 - Issue 4 - p 279-283


The kneeling prone position, also called the Wiltse modification of the Andrews frame, is often used for low back surgery. The patient's abdomen is free, there is a sternal support, and the knees are flexed. The position increases the exposure of the vertebral canal [1] and may decrease intraoperative bleeding [2]. Intra-operative bleeding is assumed to be decreased because venous pressure is decreased which in turn is caused by pooling of blood in the extremities and in the abdomen which has no external support. A concern for the anaesthesiologist is whether or not regional blood perfusion is sufficient in the kneeling prone position. If not, how do we detect inadequate perfusion? Total body metabolism should not change because the body position is changed. Thus, a decrease in oxygen uptake rate or in carbon dioxide elimination may be indicators of hypoperfusion. Body positioning and its effects on respiratory gas exchange are not extensively investigated, and most data concern the classical horizontal prone position.

In normal awake subjects, the classical horizontal prone position does markedly affect respiratory function [3]. This position has been efficient in improving oxygenation in some patients with adult respiratory distress syndrome (ARDS) [4,5], and also in normal mechanically ventilated patients [6]. In another study on healthy patients undergoing elective surgery, oxygenation and functional residual capacity (FRC) were improved 20 min after adoption of the prone position [7]. Increased oxygenation is also seen without increased FRC [6]. Different explanations for the increased oxygenation without changes in FRC include more uniformly distributed lung perfusion [8], and improved ventilation and perfusion (V/Q) matching [9].

In our study we examined whether the increased oxygenation, seen in the prone position, is present immediately after adoption of this position was completed, and if the prone position alters oxygen consumption (VO2) and carbon dioxide excretion (VCO2).


The study was approved by the local Hospital Ethics Committee. After informed verbal consent, 30 ASA I-II patients scheduled for low back surgery in the kneeling prone position were studied. All patients were free from respiratory problems, except one patient with mild, well-controlled asthma. Two patients had mild hypertension. Flunitrazepam 0.5-1 mg orally was given for premedication according to the routine of the ward.

Anaesthesia was induced using fentanyl 3 μg kg−1 and thiopental 4-5 mg kg−1; pancuronium 0.1 mg kg−1 was used for muscle relaxation. When needed, fentanyl and pancuronium were repeated.

After endotracheal intubation, the patient's lungs were ventilated by a Siemens Servo® 900C nonrebreathing system (Siemens Elema, Solna, Sweden). The standardized induction procedure was completed within 10 min. Anaesthesia was maintained with isoflurane with an end-tidal (ET) concentration of 1.2%. Inspired oxygen fraction (FiO2), was set to 0.4 and normoventilation was attained by maintaining end-tidal CO2 at 5.0 ± 0.3 vol%. The ventilator was set to a frequency of 12 breaths min−1, constant flow, inspiration time 25%, plateau 10%. No positive end-expiratory pressure or recruitment manoeuvres were used. Cuff pressure was set to 30 cmH2O. The operating table used was the Andrews Spinal Surgery Table SST-3000® (Orthopedic Systems Inc., Hayward, CA, USA). Once induction was completed, arterial and central venous catheters were inserted. After 10 min of lung ventilation, when stable anaesthesia with an end-tidal isoflurane concentration of 1.2% had been reached, the physiological measurements with the patient in the supine position were performed. Then, the patient was placed in the kneeling prone position, which was completed within another 10 min. New arterial blood-gases were taken 1 min after completion of the positioning procedure. During the study period body temperature was not recorded, but measures were taken to avoid any temperature falls, by warming the operating theatre and administer warm intravenous fluids.

For studies of gas exchange, a Datex Ultima® SV (Datex Instrumentarium OY, Helsinki, Finland) was used to register FiO2, end-tidal oxygen fraction (FETO2), end-tidal carbon dioxide fraction (FETCO2), inspired carbon dioxide fraction (FiCO2), mixed expired carbon dioxide fraction (FECO2), inspired oxygen fraction (FiO2), mixed expired oxygen fraction (FEO2) and expired minute volume (VE). The tidal volume measurements have an accuracy of ±6% within the interval 250-2000 mL. The gas analyses were recorded on line to a computer using a Datex Daisy® software program. Mixed expired gas was sampled from a 2 L mixing box connected distally to the expiratory outlet valve of the ventilator.

Inspiratory and end-tidal gas samples were taken from a port between the Y-piece of the breathing system and the endotracheal tube. Carbon dioxide fraction and expired minute volume ventilation, i.e. FECO VE Oxygen uptake rate, VO2, was calculated from the difference between inspired (FiO2VI), and expired (FEO2 VE) oxygen content. Haldane transformation was used for calculation of inspired minute ventilation (VI):

As this transformation is done with the assumption that there is no net exchange of inert gas, corrections were made for uptake of isoflurane and elimination of nitrogen. For this purpose, the following formulae were used:

• Isoflurane uptake rate: ETisofl1.2% (mL min−1 70 kg−1): 81t−0.5[10];

• Nitrogen elimination rate: FiO2 0.4 (mL min−1 70 kg−1): (79-58)/79 100t−0.5[11];

• Alveolar ventilation (VA, L m−2min−1) was calculated through the equation VA = kVCO2 (L m−2 min−1)/FETCO2.

The constant adjusts for gas volumes at body temperature and pressure saturated (BTPS) and is dependent on room temperature and barometric pressure (k = 1.17-1.19) [12].

Cardiac output was monitored non-invasively by measuring transthoracic bioimpedance, using NCCOM3-R7® (BioMed Medical Manufacturing Ltd, Irvine, CA, USA) [13]. A Datex Cardiocap® (Datex Instrumentarium OY, Helsinki, Finland) was used for measurements of heart rate and mean arterial pressure.


Statistical analyses were performed by comparing the variables measured in the supine position (supine) - immediately before turning the patient into the kneeling prone position with the values measured at 5 and 10 min after completion of the positioning procedure (kneeling prone position; KPP 5 min and KPP 10 min, respectively), except for arterial blood-gases where the supine variables were compared with variables measured 1 min after completion of the positioning procedure.

Median values and 25-75 percentiles are given, if not stated otherwise. Within group comparisons were made by repeated measures ANOVA followed by Wilcoxon's signed rank sum test. Statistical significance was assumed for values of P < 0.05.


The patients had a median age of 40 yr (range 24-67), weight of 79 kg (range 54-107), height of 177.5 cm (range 156-202 cm) and haemoglobin concentration of 146 g L−1 (range 110-167 g L−1). There were 13 female patients.

The median expiratory ventilation volume was 5.0 L min−1 and was not significantly affected by positioning the patients in the kneeling prone position (Table 1). The oxygen uptake rate did not significantly change compared with the supine position of 76 mL min−1 m−2. The carbon dioxide excretion rate decreased from a supine position value of 71 mL min−1 m−2 to 66 mL min−1 m−2 at 5 min (P < 0.001) and 10 min (P < 0.01) after completion of the kneeling prone position.

Table 1
Table 1:
Expiratory ventilation, oxygen uptake and carbon dioxide excretion.

The alveolar ventilation (VA), was 1.76 (1.63-1.90) L min−1 m−2 in the supine position. After completion of the kneeling prone positioning, VA decreased to 1.62 (1.45-1.78) L min−1 m−2 at 5 min (P < 0.001) and 1.67 (1.50-1.78) L min−1m−2 at 10 min (P < 0.01). The end-expired to arterial oxygen partial pressure difference (PETO2 − PaO2) significantly decreased in the kneeling prone position. The median value was 11.1 (7.0-18.7) kPa in supine position compared with 7.8 (4.1-10.8) kPa in the prone position (P < 0.001).

In association with turning the patient into the prone position there were significant rises in PaO2 (P < 0.001) (Fig. 1). The SaO2 increased significantly from a median value of 99.2% to 99.5% in the kneeling prone position (P < 0.001).

Figure 1
Figure 1:
Oxygenation in the supine and kneeling prone positions during anaesthesia. Oxygenation (PaO2, kPa) during supine position and after turning the patient into the kneeling prone position. The two interrupted vertical lines represent median values (25-75 percentiles) in the supine position, 21.1 (13.2-25.8), and in the kneeling prone position 25.3 (22.1-29.0). *P < 0.001 for the difference between supine and the kneeling prone position. Filled bars represent changes in oxygenation in each of the 30 patients when turning the patient into the kneeling prone position. All except two patients (Nos 18 and 24) increased their oxygenation.

Heart rate significantly increased (P < 0.001), and cardiac output significantly decreased (P < 0.001) in association with adoption of the kneeling prone position (Table 2). Mean arterial pressure and central venous pressure were unaltered.

Table 2
Table 2:
Circulatory changes.


Indirect calorimetry, although recommended as the best available method, has some limitations in the measurements of VO2. FiO2 exceeding 0.5 and the presence of anaesthetic inhalation agents cause errors in the measurement [14]. To compensate for these effects, we kept the FiO2 at 0.4 and corrected the oxygen uptake rates for anticipated rates of isoflurane uptake and nitrogen elimination.

The present study demonstrates that the kneeling prone position has little effect on oxygen consumption, but immediately improves oxygenation as measured by arterial oxygen saturation, and arterial oxygen tension. On the other hand, alveolar ventilation decreased. A possible explanation for this decrease could be found in the calculation of alveolar ventilation where VCO2 is a main factor (VA = kVCO2/FETCO2). The reduced VA could reflect a reduced CO2 elimination as was seen in our study after the patients were placed in the kneeling prone position. The expired ventilation volume and the end-tidal CO2 concentration were unchanged, but the mixed expired CO2 fraction decreased in association with positioning, which explains the decrease in VA. In the kneeling prone position, the legs are positioned below the level of the heart and pooling of blood in the legs may well result in decreased cardiac output. This may well have contributed to reduced CO2 elimination and, therefore, reduced calculated alveolar ventilation. It is rather unlikely that the reduced CO2 elimination in the kneeling prone position was a reflection of reduced CO2 production. It is more likely an indication that the carbon dioxide transport to the lungs, i.e. cardiac output, was reduced. During acute changes in cardiac output, there is a direct relationship between cardiac output and end-tidal CO2 tension [15,16]. This phenomenon is transient and in our study end-tidal CO2 was unchanged despite a moderate reduction in cardiac output.

Our results are in part in agreement with a study by Pelosi and colleagues who found a marked increase in PaO2 in healthy anaesthetized orthopaedic patients 20 min after turning them from the supine to the prone position [7]. The results of our study indicate that improvement in oxygenation is fast in onset, and is present directly after the completed re-positioning.

Using a prone position with chest and sternum support, but not with dependant lower extremities as in the kneeling prone position, Pelosi and colleagues also found a marked increase in FRC. Increased FRC is also seen in awake subjects in the horizontal prone position [17]. On the other hand, improved PaO2 in the prone position has been seen without changes in FRC [6,18]. Further explanation for the improved oxygenation in the prone position may be an improvement of the V/Q ratio in the lungs [8], and a more uniform distribution of perfusion [9]. Although we did not measure FRC in this study, V/Q ratio or any other parameter that could explain improved oxygenation in the prone position, our results are most likely due to the ventilatory effects of the prone position. Other studies have focused on detecting changes in oxygenation later than in the present study, e.g. 15 min [19] and 20 min [7], after completing the prone position. The main objective of the present study was to ascertain if an increased oxygenation can be detected directly after adoption of the kneeling prone position has been completed.

In conclusion, we find that mechanisms involved in increased oxygenation in the prone position are fast in onset, but the prone position itself has no increased demands on oxygen consumption. We cannot exclude the possibility that the calculated decreased alveolar ventilation in our study is dependent on circulatory changes, and further studies are needed to clarify the results.


Financial support was provided by grants from the Medical Faculty of University of Gothenburg and Kommunförbundet Västra Götaland, Sweden.


1. Ecker A. Kneeling position for operations on the lumbar spine. Especially for protruded intervertebral disc. Surgery 1949; 25: 112-113.
2. Böstman O, Hyrkäs J, Hirvensalo E, Kallio E. Blood loss, operating time, and positioning of the patient in lumbar disc surgery. Spine 1990; 15: 360-363.
3. Lumb AB, Nunn JF. Respiratory function and rib cage contribution to ventilation in body positions commonly used during anaesthesia. Anesth Analg 1991; 73: 422-426.
4. Jolliet P, Bulpa P, Chevrolet JC. Effects of the prone position on gas exchange and hemodynamics in severe acute respiratory distress syndrome. Crit Care Med 1998; 26: 1977-1985.
5. Flaaten H, Aardal S, Hevroy O. Improved oxygenation using the prone position in patients with ARDS. Acta Anaesthesiol Scand 1998; 42: 329-334.
6. Brazzi L, Ravagnan I, Pelosi P, Gattinoni L. Prone position in anaesthesia and intensive care. Care Crit Ill 1999; 15: 5-10.
7. Pelosi P, Croci M, Calappi E, et al. The prone positioning during general anesthesia minimally affects respiratory mechanics while improving functional residual capacity and increasing oxygen tension. Anesth Analg 1995; 80: 955-960.
8. Nyrén S, Mure M, Jacobsson H, Larsson SA, Lindahl SGE. Pulmonary perfusion is more uniform in prone than in supine position: scintigraphy in healthy humans. J Appl Physiol 1999; 86: 1135-1141.
9. Mure M, Lindahl SGE. Prone position improves gas exchange - but how? Acta Anaesthesiol Scand 2001; 45: 150-159.
10. Bengtsson JP, Bengtsson A, Stenqvist O. Uptake of enflurane and isoflurane during spontaneous and controlled ventilation. Anaesth Intens Care 1992; 20: 191-195.
11. Bengtsson JP, Bengtsson J, Bengtsson A, Stenqvist O. Sampled gas need not be returned during low flow anesthesia. J Clin Monitor 1993; 9: 330-334.
12. Nunn JF. Applied Respiratory Physiology, 4th edn. Oxford, UK: Butterworth-Heinemann, 1993: 128-129, 195-197.
13. Thomas SH. Impedance cardiography using the Sramek-Bernstein method: accuracy and variability at rest and during exercise. Br J Clin Pharmacol 1992; 34: 467-476.
14. Aukburg SJ, Geer RT, Wollman H, Neufeld GR. Errors in measurement of oxygen uptake due to anesthetic gases. Anesthesiology 1985; 62: 54-59.
15. Shibutani K, Muraoka M, Shirasaki S, Kubal K, Sanchala VT, Gupte P. Do changes in end-tidal PCO2 quantitatively reflect changes in cardiac output? Anesth Analg 1994; 79: 829-833.
16. Wahba RWM, Tessler MJ, Béïque F, Kleiman SJ. Changes in PCO2 with acute changes in cardiac index. Can J Anesth 1996; 43: 243-245.
17. Mahajan RP, Hennessy N, Aitkenhead AR, Jellinek D. Effect of three different surgical prone positions on lung volumes in healthy volunteers. Anaesthesia 1994; 49: 583-586.
18. Numa AH, Hammer J, Newth CJ. Effect of prone and supine position on functional residual capacity, oxygenation, and respiratory mechanics in ventilated infants and children. Am J Respir Crit Care Med 1997; 156: 1185-1189.
19. Mutoh T, Guest RJ, Lamm WJE, Albert RK. Prone position alters the effect of volume overload on regional pleural pressures and improves hypoxemia in pigs in vivo. Am Rev Respir Dis 1992; 146: 300-306.

BLOOD GAS ANALYSIS, oximetry; GASES, carbon dioxide; METABOLISM, oxygen consumption; POSTURE, prone position

© 2004 European Academy of Anaesthesiology