EFFECT OF OXYGEN THERAPY ON LATE POSTOPERATIVE EPISODIC AND CONSTANT HYPOXAEMIA

British Journal of Anaesthesia 1992; 68: 18-22

EFFECT OF OXYGEN THERAPY ON LATE POSTOPERATIVE EPISODIC AND CONSTANT HYPOXAEMIA J. ROSENBERG, M. H. PEDERSEN, P. GEBUHR AND H. KEHLET SUMMARY

KEY WORDS Complications: postoperative hypoxaemia Oxygen: therapy.

Arterial hypoxaemia is common as a constant or episodic phenomenon in the postanaesthesia recovery room [ 1 ] and in the surgical ward [2, 3] after major abdominal or orthopaedic surgery. Episodic hypoxaemia is more frequent on the second than on the first night after operation [2, 4]. Our preliminary findings suggest that episodic nocturnal postoperative hypoxaemia may be related to increased cardiac demands [2, 4], and occurs despite oxygen therapy [2]. As constant hypoxaemia itself may trigger the development of apnoea and periodic breathing [5], we have examined the effect of oxygen therapy on the occurrence and severity of late postoperative episodic and constant hypoxaemia on the second night after hip surgery.

The study was approved by the local Ethics Committee. We studied 40 consecutive patients undergoing elective total hip replacement (table I). The patients received either 21 % or 37 % oxygen by Hudson mask on the second night after operation (23:00 to 07:00); no oxygen was given on the first night. Using a variable performance Hudson mask with a flow of oxygen 3 litre min"1 and air 12 litre min"1, the oxygen concentration flowing into the mask is 37 %, but the effective oxygen concentration inspired by the patient may be less if inspiratory flow exceeds air flow, resulting in increased mixture with room air [6, 7]. We did not choose a Venturi mask partly because it is not the standard in our hospital and partly because it gives an unacceptable audible noise [8]. Oxygen saturation was measured and stored in the internal memory of a Nellcor N-200 (software version 2.7) pulse oximeter using an adhesive finger probe [9], with subsequent data printout on a chart recorder. The oximeter memory stores up to 12 h of oxygen saturation data, sampled once every 1 s and averaged every 5 s. The pulse oximetry printouts were analysed for mean all-night oxygen saturation by estimating average values for every 15-min period throughout the recording, followed by averaging of all the periods. Sudden hypoxaemia was denned as a decrease in oxygen saturation of 5 % or more within 2 min. These desaturations were classified into those with an endpoint equal to or greater than 80 % oxygen saturation and those with an end-point less than 80%. Minimum oxygen saturation was defined as the minimum single value for oxygen saturation obtained during the study. The study was a controlled, randomized (by envelope), double-blind design (patient blind to oxygen regimen during the treatment period, and observer blind to oxygen regimen during data analysis). Exclusion criteria included pre-existing symptoms of neurological, cardiac or respiratory disease, including excessive daytime sleepiness. Data were excluded from analysis if the patient had worn the mask for less than 7 h of the 8-h observation JACOB ROSENBERG, M.D., MDCAEL HJORTH PEDERSEN, M.D., HENRIK

KEHLET, M.D., PH.D. (Department of Surgical Gastroenterology); PETER GEBUHR, M.D. (Department of Orthopaedic Surgery);

Hvidovre University Hospital, DK-2650 Hvidovre, Denmark. Accepted for Publication: July 31, 1991. Correspondence to J.R.

Downloaded from http://bja.oxfordjournals.org/ at University of Aberdeen on March 21, 2015

As constant hypoxaemia itself may trigger development of apnoea and periodic breathing, we have studied the effect of oxygen therapy on the occurrence of late postoperative episodic hypoxaemia. Thirty-five patients without cardiopulmonary disease and undergoing elective total hip replacement were monitored with a pulse oximeter on the second night after operation (23:00 to 07:00), receiving either 21% or 37% oxygen by face mask in a randomized double-blind design. Mean oxygen saturation was greater in the group receiving 37% oxygen than in those having 21% oxygen (96% vs 92%, ?<0.01). There was a weak correlation between mean oxygen saturation and the total number of hypoxaemic episodes (x, = -0.62, P < 0.001), explained partly by the calculated (non-mechanistic) reduction in mean saturation by the episodes of hypoxaemia. There was no significant difference between the groups in the total number of sudden decreases in oxygen saturation, the duration of the events or number of patients with events to more or less than 80% oxygen saturation, although there was a trend towards fewer patients having events to less than 80% in the 37% oxygen group (nine of 17patients vs five of 18 patients (ns); 95 % confidence limits of median difference: —6 to 56%). We conclude that postoperative oxygen therapy with 37% oxygen by face mask increases mean oxygen saturation, but does not influence the basic mechanism leading to episodic hypoxaemia.

PATIENTS AND METHODS

OXYGEN THERAPY AND EPISODIC HYPOXAEMIA

19

TABLE I. Patient characteristics, intraoperative data and postoperative Confidence limits are calculated according to standard formulae [10,11]; between groups

(Pc)

Group 2 (37% O.) (P.)

70 (50-92) 165(149-180) 71(56-85) 900(100-1800) 90(55-145) 0.4(0.3-0.7)

70(42-107) 165(152-183) 74 (54-32) 750(150-2200) 105 (65-145) 0.4 (0.2-0.5)

Group 1 (21 %O.) No. patients Weight (kg) Height (cm) Age (yr) Blood loss (ml) Duration of surgery (min) Intraoperative fentanyl (mg) Opioid admin, (mg) Day 0 Day 1 Total

opioid administration {median ^ P = (Pc — Pt). No significant

17

{range)). differences

95 % confidence limits of median difference!

18

14(0-80) 10(0-60) 28 (0-140)

13 (0-50) 10(0-^H)) 24(0-88)

-6 -5 -7 -250 -30 -0.05

< P < 12
- 9 < P < 10 - 5 < P < 10 -12 < P < 2 0

Group 1 (21% O.) Oxygen satn (%) Mean Minimum Total number of episodes No. patients with episodes No. episodes at > 80 % No. patients with episodes at 2=80% No. episodes at < 80 % No. patients with episodes at <80% No. episodes < 1 min No. episodes 1-5 nun No. episodes > 5 min

(Pc)

Group 2 (37% O.) (P,)

95 % confidence limits of median difference f

92 (83-96) 78 (62-91) 29 (2-323) 17 of 17 (= 100%) 20 (2-293) 17 of 17 (= 100%)

96 (89-99)** 86 (65-94) 8(0-207) 16 of 18 (=89 %) 6(0-204) 16 of 18 (=89%)

- 5 . 5 < P < -1.0 -11 < P < 0 - 2 < P < 29 -4%
1 (0-30) 9 of 17 (=53%)

0(0-23) 5 of 18 (=28%)

0
4(0-184) 4(0-^6) 0(0-1)

period, and if less than 7 h of pulse oximetry data were obtained on the observation night (because of accidental oximeter disconnection, etc.). The nursing staff maintained records of the mask position on an hourly basis throughout the night. All patients received diazepam 5-10 mg for premedication; anaesthesia was induced and maintained with thiopentone, midazolam, fentanyl, suxamethonium, pancuronium and nitrous oxide in oxygen. Postoperative analgesia comprised morphine or ketobemidone 5-10 mg i.m. on demand; opioid administration was monitored for the first 48 h after surgery. Sedative drugs were not allowed in the postoperative period. Data were analysed using the Mann-Whitney test, Fisher exact test and Spearman's rank correlation test as appropriate. The level of significance was P < 0.05. RESULTS

Data from two of the 40 patients studied were excluded because of displacement of the oxygen mask for more than 1 h of the night, and in three patients data were missing; thus there were data from 35 patients for analyses. The two patient groups were comparable in weight, height and age, and intraoperative and postoperative variables other than oximetry data (table I). The group receiving 37 % oxygen treatment

18 (2-286) 5(0-37) 0(0-1)

0 < P < 24 - 2 < P < 12 0< P<0

100 -i

80-

02:00

02:30 Time

03:00

FIG. 1. Changes in oxygen saturation (Spo,) on the second night after hip surgery in a 75-yr-old woman receiving 37 % oxygen by face mask.

had less constant hypoxaemia than the group receiving 21 % oxygen treatment (P < 0.01) (table II). There was no difference between groups in the total number of sudden decreases in oxygen saturation (ns) and number of events lasting less than 1 min, 1-5 min and more than 5 min (ns). Figure 1 shows the changes in oxygen saturation in a patient receiving 37 % oxygen. Six patients in group 1 and five patients in group 2 had more than five episodes of sudden desaturation per hour of monitoring. For all patients (n = 35), we found a statistically significant but weak correlation between the mean oxygen saturation and the total number of hypox-

Downloaded from http://bja.oxfordjournals.org/ at University of Aberdeen on March 21, 2015

TABLE II. Postoperative oxygen saturation results (median {range)). Confidence limits are calculated according to standard formulae [11),ll];-fP = Pc-P,. **P < 0 01 compared with group 1

BRITISH JOURNAL OF ANAESTHESIA

20 100

r, = -0.62 P< 0.001

80100

200

300

400

No. episodes FIG. 2. Correlation between mean oxygen saturation (SpOt) and total number of hypoxacmic episodes on the second night after hip surgery, x = 21 % oxygen group; • = 37% oxygen group.

DISCUSSION

We have found that, whilst postoperative oxygen therapy increased mean oxygen saturation, it did not alter the occurrence of sudden episodes of desaturation. For practical reasons we did not choose to monitor a preoperative night, therefore we may have overlooked patients suffering from sleep apnoea syndrome before surgery, although none reported increased daytime sleepiness before operation. Nevertheless, the results showed that the incidence of severely disordered breathing was comparable in the two groups: six patients in group 1 and five patients in group 2 had more than five episodes of sudden desaturation per hour of monitoring. Further-

TABLE III. Influence of t.m. opioid analgesics on late postoperaltve episodic hypoxaemia. Only eight of the 35 patients [presented in the table) received opioid analgesic during the study night. Analgesic doses are presented as morphine equivalents. — = No episode of hypoxaemia after analgesic No. of episodes Patient no.

No. doses

Dose (mg)

Time analgesia to first hypoxaemia (min)

1 2 3

1 1 2

4 5 6 7

1 1 1 2

8

1

10 10 5 5 10 2.5 5 10 10 10

40 — 15 — 55 — 20 10 15 —

=£2h before analgesic

>2h after analgesic

Treatment group (% oxygen)

9 8 0 0 3 16 65 2 6 3

8 0 7 0 11 0 36 6 2 0

21% 21% 21% 37% 37% 37 ° o 21% 37%

Downloaded from http://bja.oxfordjournals.org/ at University of Aberdeen on March 21, 2015

aemic episodes (rs = - 0 . 6 2 , P < 0.001) (fig. 2). There was no correlation between amount of opioid used in the 24-h period before the observation night and the number of sudden desaturations (ns). Only eight of the 35 patients received opioid injections during the study night. The influence of opioid analgesics on late postoperative episodic hypoxaemia in these eight patients is shown in table III. Three patients did not receive any analgesics during the entire postoperative period, but quality of analgesia was not monitored.

more, our study resembles the clinical situation in which no preoperative monitoring is performed and postoperative oxygen therapy is prescribed in order to reduce the incidence of hypoxaemia. Earlier studies have suggested a detrimental effect of postoperative morphine analgesia on ventilatory pattern. After upper abdominal surgery the amount of opioid has been shown to correlate with the degree of respiratory depression [12], and i.v. morphine has been shown to produce a greater number of incidents of apnoea and episodic hypoxaemia compared with extradural local anaesthetic techniques [1, 13]. The present study did not confirm these findings. As demonstrated in table III, only eight of the 35 patients received opioids during the study night. In four of these eight patients there was no episode of hypoxaemia after the injections, and in the remaining four patients there was no constant pattern of distribution of the episodes of hypoxaemia before and after the opioid injections. Furthermore, there was no statistically significant overall correlation between the amount of opioid used in the period 24 h before the study night and the number of sudden desaturations. Thus the effect of i.m. morphine analgesia on the development of late postoperative episodic hypoxaemia is still unclear. There is increasing evidence that the development of episodic hypoxaemia may be triggered by constant hypoxaemia itself. Thus periodic breathing is well documented during sleep at high altitude [14, 15]. The respiratory pauses during sleep at high altitude are presumably of central origin; they are not associated usually with snoring or any other suggestion of obstruction and are accompanied by absence of rib cage and abdominal activity [5]. This respiratory arrhythmia seems to arise from the combination of hypocapnia, which decreases respiratory drive, hypoxaemia, which stimulates the termination of apnoea and the occurrence of hyperpnoea with consequent hypocapnia, leading to perpetuation of periodicity [5]. However, the basic physiological mechanism remains to be clarified [15]. Our finding of a significant but weak correlation between degree of constant hypoxaemia and occurrence of episodic hypoxaemia may support the findings in the high altitude studies. However, the correlation may be partly explained by the calculated (non-mechanistic)

OXYGEN THERAPY AND EPISODIC HYPOXAEMIA

desaturations [25]. However, the present study failed to show any effect of oxygen therapy on the duration and the number of the sudden desaturations in the surgical setting. We have shown, not surprisingly, that late postoperative nocturnal constant hypoxaemia may be improved with oxygen therapy. Constant hypoxaemia may correlate with local subcutaneous oxygen tension in elective surgical patients [27] and may subsequently reduce synthesis of collagen in the surgical wound [28] and reduce resistance to bacterial wound infections [29]. Furthermore, constant hypoxaemia and episodic changes in oxygen saturation may contribute to development of postoperative myocardial ischaemia and infarction [4, 30]. Therefore, oxygen therapy may be of major importance for patient care in both the early and later postoperative periods. However, the level of endangering hypoxaemia still has to be settled for different kinds of surgery and in different patient categories, and studies on possible therapeutic measures against the postoperative ventilatory dysfunction are needed to rationalize postoperative oxygen therapy. ACKNOWLEDGEMENTS The study was supported by The Danish Medical Research Council (12-8033), The Danish Heart Foundation, P. Carl Petersen Foundation, Toyota Foundation Denmark, Velux Foundation and the Foundation of 17-12-1981. REFERENCES 1. Catley DM, Thorton C, Jordan C, Lehane JR, Royston D, Jones JG. Pronounced episodic oxygen desaturation in the postoperative period: its association with ventilatory pattern and analgesic regimen. Anesthesiology 1985; 63: 20-28. 2. Rosenberg J, Dirkes WE, Kehlet H. Episodic arterial oxygen desaturation and heart rate variations following major abdominal surgery. British Journal of Anaesthesia 1989; 63: 651-654. 3. Martin VC. Hypoxaemia in elderly patients suffering from fractured neck of femur. Anaesthesia 1977; 32: 852-867. 4. Rosenberg J, Rasmusscn V, Jessen FV, Ullstad T, Kehlet H. Late postoperative episodic and constant hypoxaemia and associated ECG-abnormalities. British Journal of Anaesthesia 1990; 65: 684-691. 5. Berssenbrugge A, Dempsey J, Iber C, Skatrud J, Wilson P. Mechanisms of hypoxia-induced periodic breathing during sleep in humanj. Journal of Physiology (London) 1983; 343 : 507-524. 6. Leigh JM. Variation in performance of oxygen therapy devices. Annals of the Royal College of Surgeons of England 1973; 52: 234-253. 7. Bethune DW, Collis JM. The evaluation of oxygen masks. Anaesthesia 1967; 22: 43-54. 8. Leigh JM. Audible noise levels of oxygen masks operating on Venturi principle. British Medical Journal 1973; 4: 652. 9. Severinghaus JW, Naifeh KH, Koh SO. Errors in 14 pulse oximeters during profound hypoxia. Journal of Clinical Monitoring 1989; 5: 72-81. 10. Campbell MJ, Gardner MJ. Calculating confidence intervals for some non-parametric analyses. British Medical Journal 1988; 296: 1454-1456. 11. Gardner MJ, Altman DG. Confidence intervals rather than P values: estimation rather than hypothesis testing. British Medical Journal 1986; 292: 746-750. 12. Catling JA, Pinto DM, Jordan C, Jones JG. Respiratory effects of analgesia after cholecystectomy: comparison of continuous and intermittent papaveretum. British Medical Journal 1980; 281: 478-^80. 13. Clybum PA, Rosen M, Vickers MD. Comparison of the

Downloaded from http://bja.oxfordjournals.org/ at University of Aberdeen on March 21, 2015

reduction in mean saturation by episodes of hypoxaemia. Furthermore, we do not know if episodes of postoperative hypoxaemia were accompanied by changes in Pco 2 , and the type of ventilatory disturbance preceding hypoxaemia has not been clarified. In the early postoperative period, the ventilatory disturbance preceding the episodes of hypoxaemia may be almost exclusively obstructive [1], whereas our own preliminary observations [unpublished] suggest that episodic hypoxaemia in the late postoperative period may be caused by different mechanisms such as central, obstructive and mixed apnoeas, hypopnoeas and periodic breathing. However, as there was no significant difference between the two groups of patients in the amount and duration of episodic hypoxaemia, it is obvious that mechanisms other than the level of constant hypoxaemia are involved in the development of episodic hypoxaemia in the late postoperative period. Thus the basic mechanism leading to episodic hypoxaemia is not radically changed with oxygen therapy. Constant hypoxaemia may be present for up to 1 week after major surgery [16, 17]. Systematic studies on the effectiveness of oxygen therapy in this late postoperative period have not been published. Constant hypoxaemia may be improved by oxygen therapy with 35 % oxygen delivered by face mask on the day after operation [18], but the hypoxaemia returns when the oxygen therapy is discontinued [19]. Only one study has previously evaluated the effect of oxygen therapy on the occurrence of episodic hypoxaemia [20]. The results showed that administration of 28 % oxygen by face mask in the first 12 h after operation abolished the sudden decreases in oxygen saturation but did not prevent the occurrence of apnoeas [20]. The present study, performed in the late postoperative period, failed to show any effect of oxygen therapy on the development of episodic hypoxaemia, while the level of constant hypoxaemia was, not unexpectedly, improved. The reason for the differing results may be a greater effect of an apnoea on arterial oxygen saturation in the late postoperative period when lung volumes are smaller than early after surgery [21], as it has been shown previously that apnoea produces a lesser end-point of arterial oxygen saturation if initial lung volume is smaller [22]. There is concern that nocturnal oxygen therapy may increase the duration of apnoea (but not the number of apnoeas), thereby leading to respiratory acidosis in patients with diseases featuring nocturnal hypoxaemia such as sleep apnoea syndromes, congestive heart failure or chronic obstructive pulmonary disease [23-25] (although intermittent respiratory acidosis in the surgical patient is not necessarily detrimental, as peripheral oxygen delivery may be improved [26]). Studies in patients with sleep apnoea syndrome have shown that short term oxygen therapy (one night) produces a slight prolongation of mean apnoea duration, but improves mean oxygen saturation [23]. Oxygen therapy in heart failure and obstructive pulmonary disease may increase mean nocturnal oxygen saturation and reduce the number of apnoeas [24] and number of

21

BRITISH JOURNAL OF ANAESTHESIA

22

14.

15. 16. 17. 18. 19. 20.

22.

23. Fletcher EC, Munafo DA. Role of nocturnal oxygen therapy in obstructive sleep apnea: when should it be used? Chest 1990; 98: 1497-1504. 24. Hanly PJ, Millar TW, Stejles DG, Baert R, Frais MA, Kryger MH. The effect of oxygen on respiration and sleep in patients with congestive heart failure. Annals of Internal Medtdne 1989; 111: 777-782. 25. Kearley R, Wynne JW, Block AJ, Boysen PG, Lindsey S, Martin C. The effect of low flow oxygen on sleep-disordered breathing and oxygen desaturation: a study of patients with chronic obstructive lung disease. Chest 1980; 78: 682-685. 26. Harken AH. The surgical significance of the oxyhemoglobin dissociation curve. Surgery, Gynecology and Obstetrics 1977; 144: 935-955. 27. Rosenberg J, Ullstad T, Larsen PN, Moesgaard F, Kehlet H. Continuous assessment of oxygen saturation and subcutaneous oxygen tension following abdominal surgery. Ada Chirurgica Scandinavica 1990; 156: 585-590. 28. Hunt TK, Conolly WB, Aronson SB, Goldstein P. Anaerobic metabolism and wound healing: An hypothesis for the initiation and cessation of collagen synthesis in wounds. American Journal of Surgery 1978; 135: 328-332. 29. Knighton DR, Fiegel VD, Halvcrson T, Schneider S, Brown T, Wells CL. Oxygen as an antibiotic: the effect of inspired oxygen on bacterial clearance. Archives of Surgery 1990; 125: 97-100. 30. Pateman JA, Hanning CD. Postoperative myocardial infarction and episodic hypoxaemia. British Journal of Anaesthesia 1989; 63: 648-650.

Downloaded from http://bja.oxfordjournals.org/ at University of Aberdeen on March 21, 2015

21.

respiratory effects of i.v. infusions of morphine and regional analgesia by extradural block. British Journal of Anaesthesia 1990; 64: 446-449. Normand H, Barragan M, Benoit O, Bailliart O, Raynaud J. Periodic breathing and O, saturation in relation to sleep stages at high altitude. Aviation Space and Environmental Medicine 1990; 61: 229-235. Ward MP, Millcdge JS, West JB. High Altitude Medicine and Physiology. London. Chapman and Hall Medical, 1989; 263-281. Knudsen J. Duration of hypoxaemia after uncomplicated upper abdominal and thoraco—abdominal operations. Anaesthesia 1970; 25: 372-377. Jones JG, Sapsford DJ, Wheatley RG. Postoperative hypoxaemia: mechanisms and time course. Anaesthesia 1990; 45: 566-573. Drummond GB, Wright DJ. Oxygen therapy after abdominal surgery. British Journal of Anaesthesia 1977; 49: 789-797. Parfrey PS, Harte PJ, Quinlan JP, Brady MP. Postoperative hypoxaemia and oxygen therapy. British Journal of Surgery 1977; 64: 390-393. Jones JG, Jordan C, Scudder C, Rocke DA, Barrowcliffe M. Episodic postoperative oxygen desaturation: the value of added oxygen. Journal of the Royal Society of Medicine 1985; 78: 1019-1022. Craig DB. Postoperative recovery of pulmonary function. Anesthesia and Analgesia 1981; 60: 46-52. Series F, Cormier Y, La Forge J. Role of lung volumes in sleep apnoea-related oxygen desaturation. European Respiratory Journal 1989;2: 26-30.