Approximately 10% of the population undergoes surgery each year, and most of these people experience postoperative pain (1). Despite a growing understanding of the basic science of pain, postoperative pain management is often inadequate (2) such that quality improvement and clinical research efforts must continue (3). Whereas the primary goal is to relieve suffering and improve quality of life after surgery, evidence suggests that improved postoperative pain management may also reduce serious morbidity and mortality (4). Physiological studies have quantified the contribution of postoperative pain to pulmonary dysfunction (5), and poorly controlled pain has been related to postoperative pulmonary (6), cardiac (7), and thromboembolic (8) complications. Given the pathogenic mechanisms of these complications (e.g., hypoventilation leading to atelectasis; decreased ambulation leading to venous stasis), it is imperative that adverse physiological sequelae related to movement-evoked pain be more effectively prevented.
Postoperative analgesic trials are increasingly including measures of pain evoked by movement and/or tactile stimulation (9,10). With this approach, treatment differences may further distinguish evoked from spontaneous pain. For example, Tverskoy et al. (11) found that doses of the opioid alfentanil, which completely suppress postoperative resting pain, are far less effective at reducing movement-evoked pain. In contrast to these findings, we have observed (12) that the novel 2-amino-3-hydroxy-5-methyl-4-isoxazole-proprionic acid/kainate receptor antagonist LY293558 produced minimal effects on spontaneous pain yet significantly reduced movement-evoked pain after surgery. Such distinctions may allow for future identification of specific analgesics or analgesic combinations that optimally reduce both spontaneous and evoked postoperative pain.
It is equally important to distinguish between evoked and spontaneous postoperative pain in terms of their deleterious physiological effects. For example, pain has long been thought to contribute to postoperative pulmonary dysfunction (13), and studies have evaluated the effects of different postoperative analgesics (14) and of different surgical approaches (15) on both pain and pulmonary function. However, the relative contributions of evoked versus spontaneous pain to pulmonary dysfunction have not been characterized. In patients undergoing abdominal hysterectomy, we have elected to use clinically relevant measures of pain evoked by ambulation (sitting up) and by cough. We have chosen peak expiratory flow (PEF) as a representative index of pulmonary function. PEF is a simple and rapid test of ventilatory capacity (16), which is effort-dependent and affected by airway obstruction and neuromuscular dysfunction (17). PEF is often used to track changes in ambulatory patients with obstructive lung disease (18) but has also been successful in distinguishing the pulmonary effects of different analgesic interventions after thoracic (19) and abdominal procedures (20). Thus, we have conducted a prospective study to examine the relationship between evoked versus spontaneous pain and peak expiratory flow after hysterectomy. The purpose of this study is to test the hypothesis that movement-evoked pain correlates with peak expiratory flow more closely than spontaneous pain in the postoperative period.
Ethics approval was obtained from the Queen’s University Research Ethics Board. We were unable to find any previous studies that directly estimated the correlation of movement-evoked pain intensity with PEF. Therefore, for the purpose of sample size estimation, we assumed an expected correlation coefficient value (r) of 0.6 with a two-tailed α value of 0.05 and a β value of 0.2, resulting in a sample size of 25 (21). Thus, this was a prospective study of 25 consenting ASA physical status I or II female patients undergoing elective total abdominal hysterectomy through a low transverse incision. Surgery was started before 12 pm (i.e., first or second case of the day). Exclusion criteria included: (a) a body mass index larger than 30 kg/m2, (b) cardiopulmonary disease/dysfunction, (c) neuromuscular disease/dysfunction, (d) persistent pain before surgery, (e) daily intake, or intake within 48 h before surgery, of any analgesic, (f) history of alcohol or substance abuse, (g) history of a major psychiatric disorder, and (h) any hypersensitivity or intolerance to opioids.
Intraoperatively, patients received a balanced anesthetic with an IV induction (propofol or sodium thiopental), muscle relaxant, inhaled volatile anesthetic, and nitrous oxide at the discretion of the attending anesthesiologist. The only intraoperative analgesics given were fentanyl 2–5 μg/kg IV at the induction followed by morphine 0.1–0.2 mg/kg IV titrated from 30 min after the induction to the end of surgery. No local or regional anesthesia or any other analgesic drugs were used during surgery. Intraoperative prophylactic antiemetic therapy included droperidol 0.5–0.625 mg IV and ondansetron 4 mg IV given 30 min before the anticipated completion of surgery. After surgery, all study patients received patient-controlled analgesia (PCA) with morphine and could also receive non-opioid co-analgesics (e.g., nonsteroidal antiinflammatory drugs) at the discretion of the attending acute pain management physician. In the postanesthetic care unit, nurses titrated morphine 2 mg IV every 5 min as required until patients were able to self-administer morphine by PCA. PCA pump settings started at a bolus dose of 1 mg, lockout interval of 5 min (no background infusion), and were adjusted as required at the discretion of the attending acute pain management physician. On the surgical ward, patients were encouraged to use an incentive spirometer five to 10 times an hour while awake and also given an abdominal splinting pillow to facilitate coughing as required. A respiratory therapist evaluated each patient every 8 h after surgery and instituted oxygen therapy for a pulse oximeter saturation reading of <92%.
At each study time point, patients completed a series of pain ratings under various conditions. These study measures, assessed in the following order, included: 100-mm visual analog scale (VAS) pain intensity at rest (REST), during sitting up in a standardized fashion from the supine position (SIT), followed by a 120-s rest period, then at maximal forced expiration (BLOW) using a peak flowmeter (Vitalograph, Ennis, Ireland), followed by a 120-s rest period, and then during a cough (COUGH). To measure BLOW pain at each time point, three maximal forced expirations were performed, and of these, only the highest VAS score (BLOW pain) and PEF were recorded. Secondary outcome measures at each time point included a resting oxygen saturation by finger pulse oximeter (Onyx, Nonin Medical Inc, Plymouth, MN) and the presence of supplemental oxygen therapy. Before surgery on the morning of the operative day, patients were asked to complete baseline pain measures and to perform three maximal forced expirations. This provided baseline data and facilitated patient teaching for the study. Pain intensity and PEF measurements were recorded on the first and second postoperative days starting at 8 am and then every 4 h until 8 pm.
Temporal changes in spontaneous and evoked pain intensity and PEF were analyzed by simple regression analysis. For purposes of data presentation and analysis, all postoperative PEFs were expressed as a percentage of baseline. Differences between spontaneous and various evoked pain measures (i.e., SIT, BLOW, and COUGH) were analyzed by two-way (pain measure by time) repeated-measures analysis of variance (ANOVA). Any main effects of pain measure were explored further using analysis of simple main effects with Fisher’s protected least significant difference test. Post hoc tests, used after significant pain measure by time interactions, were performed using Newman-Keuls multiple comparisons. In a similar fashion, differences in pain scores between patients receiving supplemental oxygen and those not receiving oxygen were analyzed by two-way (oxygen treatment group by time) ANOVA. At each separate time point (i.e., 20, 24, 28, 32, 44, 48, 52, and 56 h after completion of surgery), VAS intensity scores of all 25 patients for each pain measure (i.e., REST, SIT, BLOW, and COUGH) were plotted against their corresponding PEFs on 32 separate plots. Pain scores were similarly plotted against their corresponding oxygen saturations for patients who had not received any supplemental oxygen therapy. Correlation analysis was performed using Pearson correlation coefficients to assess the relationship between each pain measure and PEF as well as the relationship between each pain measure and oxygen saturation at each time point. All statistical tests were considered significant if they met a threshold of P < 0.05.
Thirty consecutive hysterectomy patients were considered for the study. Two patients were excluded because of chronic analgesic use, one was excluded because of chronic alcoholism, and two patients were withdrawn from the study after their planned low transverse incision was converted to a longitudinal midline incision. Thus, 25 patients completed the study. Mean (sd) age, weight, and height were 47.4 (9.1) yr, 72.3 (11.9) kg, and 160.8 (6.8) cm, respectively. Mean (sd) IV morphine consumption from the end of surgery to 8 pm on the second postoperative day was 67.3 (57.1) mg. Thirteen of the 25 study patients received a nonsteroidal antiinflammatory drug at some point during the perioperative period. For the entire study group, mean (sd) oxygen saturation across the first 2 postoperative days was 96.2% (2.1%). Nine of the 25 study patients received supplemental oxygen therapy at some point during the first 2 postoperative days. Figure 1 plots the temporal profiles of REST, SIT, BLOW, and COUGH pain as well as PEF across the first 2 postoperative days. Simple regression analysis demonstrated that all pain measures significantly diminished (P < 0.001), and PEF significantly improved (P < 0.001), across the duration of the study period. Two-way repeated-measures ANOVA resulted in a significant effect of pain measure (F = 5.28;P = 0.002) as well as a significant effect of time (F = 37.91;P < 0.001) with no significant pain-measure/time interaction (F = 1.051;P > 0.05). Fisher’s protected least significant difference test resulted in significant differences (Fig. 2) between REST and SIT pain (mean difference, 10.99;P = 0.007), REST and COUGH pain (mean difference, 15.67;P < 0.001), and BLOW and COUGH pain (mean difference, 9.39;P = 0.03). Mean VAS pain intensity (se) for REST, SIT, BLOW, and COUGH averaged over the first 2 postoperative days was 10.5 (0.8) mm, 21.5 (1.5) mm, 16.8 (1.3) mm, and 26.1 (1.7) mm, respectively (Fig. 2). Correlation analysis resulted in significant negative correlations between PEF and pain at all eight studied time points for COUGH pain, at seven time points for SIT pain, at four time points for BLOW pain, and at two time points for REST pain (Table 1). As an illustrative example, Figure 3 shows scatter plots of PEF against REST, SIT, BLOW, and COUGH pain at 20 h after completion of surgery. No significant correlations were observed between pain and oxygen saturation (data not shown). There were no differences in any of the pain measures on comparing patients receiving supplemental oxygen to those not receiving oxygen (data not shown).
The above results confirm previous evidence that during the first two days after abdominal surgery spontaneous and evoked pain diminish (22) and postoperative reductions in pulmonary function improve (13). Furthermore, pain evoked by sitting up and by coughing is significantly more severe than pain at rest. Also, pain evoked by coughing is significantly more severe than pain evoked by maximal forced expiration. Finally, pain was significantly, and negatively, correlated with PEF most frequently for COUGH and SIT pain and least frequently for REST and BLOW pain.
Evidence suggests that mechanisms underlying evoked pain are different from those underlying pain at rest (23). Inflammatory mediators released in the periphery after tissue injury, such as bradykinin, neuropeptides, serotonin, and histamine, activate A-δ and C nociceptive fibers and continuously transduce pain signals resulting in spontaneous or stimulus-independent pain (24). However, in the setting of surgical tissue injury, several more complex events occur in the peripheral and central nervous systems, resulting in stimulus-evoked pain, which is characterized by allodynia (nociception after normally innocuous stimuli) and/or hyperalgesia (enhanced nociception after a graded noxious stimulus). In addition to directly activating pain fibers, several inflammatory mediators cause peripheral sensitization, which is an alteration in the response properties of peripheral nerves such that their threshold for activation is decreased, i.e., more pain sensitive (25). In the setting of peripheral sensitization, primary afferent nociceptors, normally only activated by high-intensity stimuli, are subsequently activated by lower intensity or even non-noxious stimuli to transmit pain. Centrally, transmission of noxious stimuli from peripheral nociceptors to the spinal cord through A-δ and C nerve fibers involves the synaptic release of neurotransmitters such as substance P and glutamate onto spinal cord neurons (26). In addition to propagating pain signals to the brain, these neurotransmitters also contribute to the development of spinal sensitization, which is an alteration in the response properties of dorsal horn neurons characterized by decreased firing thresholds, as well as temporal and spatial summation and resulting in primary (at site of injury) and secondary (adjacent to site of injury) mechanical hyperalgesia (27,28). Another important feature of central sensitization is an alteration in sensory processing such that pain is pathologically transmitted via A-β fibers that normally only transmit touch (29). The consequences of all these central changes include mechanical hyperalgesia (both primary and secondary) and allodynia, although their inciting mechanisms are different from those operant in peripheral sensitization.
There are several reasons why pain evoked by sitting and by coughing are more intense than pain at rest. Possible explanations include a preferential effect of postoperatively used opioids on mechanisms of spontaneous pain (11) and/or the recruitment of more nociceptive sensory inputs during movement. Regarding the latter explanation, electrophysiological studies from a rat model of postsurgical pain demonstrate that surgical incision produces sensitization of peripheral mechanosensitive afferent nerve fibers (30) as well as sensitization of spinal dorsal horn wide dynamic range neurons centrally (31). In the present study, muscle contraction and movement of tissues (e.g., skin and peritoneum) that were directly injured during surgery (in the case of primary hyperalgesia), as well as adjacent tissues (in the case of secondary hyperalgesia), would lead to a greater nociceptive barrage compared with that which occurs at rest. Subsequently, movement-evoked pain can lead to kinesophobia (32), which, in the postoperative setting, hinders progressive ambulation and recovery of lung function.
The observed negative correlations between pain and PEF builds on direct evidence that pain adversely affects pulmonary function (5). Although the frequency of coughing was not measured in this study, we hypothesize that the consistent, significant correlation of cough-evoked pain with a reduction in PEF is, in part, due to patient avoidance of coughing, which ultimately limits deep inspiration, lung reexpansion, and clearance of secretions (33). The results of this study must be interpreted in context of the limitations of PEF as a test of lung function. It should be noted that PEF measures active expiration, which does not play a role in normal breathing (17). Therefore, future studies using other surrogates of pulmonary function (e.g., functional residual capacity) would provide a more detailed and physiologically relevant characterization of the effect of evoked pain on postoperative lung function, and larger sample sizes would be required to correlate evoked versus spontaneous pain with major outcomes such as atelectasis or pneumonia. A meta-analysis of epidural analgesia trials suggests that epidural analgesia may reduce postoperative morbidity but that measures of pulmonary function, such as forced expiratory volume in one second of expiration, forced vital capacity, and PEF, do not necessarily predict adverse respiratory outcomes (34). However, data from this study provide the first evidence that movement-evoked postoperative pain plays a particularly important role in postoperative lung dysfunction and requires further attention. Future research should be directed, first, at understanding further the unique neurophysiology of postoperative evoked pain and, second, at characterizing in more detail the physiological implications of evoked pain as they pertain to relevant postoperative respiratory, cardiac, and thromboembolic complications.
The authors wish to thank Allan Bell for his technical support and Drs Audun Stubhaug, Joel Katz, and Allison Froese for thoughtful comments made on previous versions of this manuscript.
1. Cousins M, Power I. Acute and postoperative pain. In: Wall PD, Melzack R, eds. Textbook of pain. 4th ed. Edinburgh, Scotland: Churchill Livingstone, 1999.
2. Abbott FV, Gray-Donald K, Sewitch MJ, et al. The prevalence of pain in hospitalized patients and resolution over six months. Pain 1992; 50: 15–28.
3. American Pain Society Quality of Care Committee. Quality improvement guidelines for the treatment of acute pain and cancer pain. JAMA 1995; 274: 1874–80.
4. Yeager MP, Glass DD, Neff RK, Brinck-Johnsen T. Epidural anesthesia and analgesia in high-risk surgical patients. Anesthesiology 1987; 66: 729–36.
5. Vassilakopoulos T, Mastora Z, Katsaounou P, et al. Contribution of pain to inspiratory muscle dysfunction after upper abdominal surgery: a randomized controlled trial. Am J Respir Crit Care Med 2000; 161: 1372–5.
6. Joris JL, Hinque VL, Laurent PE, et al. Pulmonary function and pain after gastroplasty performed via laparotomy or laparoscopy in morbidly obese patients. Br J Anaesth 1998; 80: 283–8.
7. Garnett RL, MacIntyre A, Lindsay P. Perioperative ischaemia in aortic surgery: combined epidural/general anesthesia and epidural analgesia vs general anesthesia and IV analgesia. Can J Anaesth 1996; 43: 769–77.
8. Modig J, Borg T, Karlstrom G, et al. Thromboembolism after total hip replacement: role of epidural and general anesthesia. Anesth Analg 1983; 62: 174–80.
9. Katz J, Clairoux M, Redahan C, et al. High dose alfentanil pre-empts pain after abdominal hysterectomy. Pain 1996; 68: 109–18.
10. Moiniche S, Dahl JB, Erichsen CJ, et al. Time course of subjective pain ratings, and wound and leg tenderness after hysterectomy. Acta Anaesthesiol Scand 1997; 41: 785–9.
11. Tverskoy M, Oren M, Dahkovsky I, Kissin I. Alfentanil dose-response relationships for relief of postoperative pain. Anesth Analg 1996; 83: 387–93.
12. Gilron I, Max MB, Lee G, et al. Effects of the 2-amino-3-hydroxy-5-methyl-4-isoxazole-proprionic acid/kainate antagonist LY293558 on spontaneous and evoked postoperative pain. Clin Pharmacol Ther 2000; 68: 320–7.
13. Craig DB. Postoperative recovery of pulmonary function. Anesth Analg 1981; 60: 46–52.
14. Tramer MR, Schneider J, Marti RA, Rifat K. Role of magnesium sulfate in postoperative analgesia. Anesthesiology 1996; 84: 340–7.
15. Ellstrom M, Olsen MF, Olsson JH, et al. Pain and pulmonary function following laparoscopic and abdominal hysterectomy: a randomized study. Acta Obstet Gynecol Scand 1998; 77: 923–8.
16. Wright BM, McKerrow CB. Maximum forced expiratory flow rate as a measure of ventilatory capacity. Br Med J 1959; 2: 1041.
17. Nunn JF. Pulmonary ventilation: mechanisms and the work of breathing. In: Nunn JF, ed. Nunn’s applied respiratory physiology. 4th ed. Oxford, UK: Butterworth-Heinemann, 1993: 117–34.
18. Kennedy DT, Chang Z, Small RE. Selection of peak flowmeters in ambulatory asthma patients: a review of the literature. Chest 1998; 114: 587–92.
19. Richardson J, Sabanathan S, Jones J, et al. A prospective, randomized comparison of preoperative and continuous balanced epidural or paravertebral bupivacaine on post-thoracotomy pain, pulmonary function and stress responses. Br J Anaesth 1999; 83: 387–92.
20. Hendolin H, Lahtinen J, Lansimies E, et al. The effect of thoracic epidural analgesia on respiratory function after cholecystectomy. Acta Anaesthesiol Scand 1987; 31: 645–51.
21. Browner WS, Black D, Newman TB, Hulley SB. Estimating sample size and power. In: Hulley SB, Cummings SR, eds. Designing clinical research. Baltimore, MD: Williams & Wilkins, 1988.
22. Tverskoy M, Oz Y, Isakson A, et al. Preemptive effect of fentanyl and ketamine on postoperative pain and wound hyperalgesia. Anesth Analg 1994; 78: 205–9.
23. Ryan SM, Watkins LR, Mayer DJ, Maier SF. Spinal pain suppression mechanisms may differ for phasic and tonic pain. Brain Res 1985; 334: 172–5.
24. Dray A. Agonists and antagonists of nociception. Proceedings of the 8th World Congress on Pain 1997:279–92.
25. Birrell GJ, McQueen DS, Iggo A, et al. PGI2-induced activation and sensitization of articular mechanonociceptors. Neurosci Lett 1991; 124: 5–8.
26. Yaksh TL, Hammond DL. Peripheral and central substrates involved in the rostral transmission of nociceptive information. Pain 1982; 13: 1–85.
27. Coderre TJ, Katz J, Vaccarino AL, Melzack R. Contribution of central neuroplasticity to pathological pain: review of clinical and experimental evidence. Pain 1993; 52: 259–85.
28. Woolf CJ, Chong MS. Preemptive analgesia: treating postoperative pain by preventing the establishment of central sensitization. Anesth Analg 1993; 77: 362–79.
29. Woolf CJ, Doubell TP. The pathophysiology of chronic pain: increased sensitivity to low threshold A beta-fibre inputs. Curr Opin Neurobiol 1994; 4: 525–34.
30. Hamalainen MM, Ozaki N, Gebhart GF, Brennan TJ. Effect of an incision on mechanosensitive afferents in the plantar rat foot. Society for Neuroscience 2000 Annual Meeting (abstract 354) 2000:3.
31. Vandermeulen EP, Brennan TJ. Alterations in ascending dorsal horn neurons by a surgical incision in the rat foot. Anesthesiology 2000; 93: 1294–302.
32. Vlaeyen JW, Kole-Snijders AM, Boeren RG, van Eek H. Fear of movement/(re)injury in chronic low back pain and its relation to behavioral performance. Pain 1995; 62: 363–72.
33. Flaminiano LE, Celli BR. Respiratory muscle testing. Clin Chest Med 2001; 22: 661–77.
© 2002 International Anesthesia Research Society
34. Ballantyne JC, Carr DB, deFerranti S, et al. The comparative effects of postoperative analgesic therapies on pulmonary outcome: cumulative meta-analyses of randomized, controlled trials. Anesth Analg 1998; 86: 598–612.