Secondary Logo

Journal Logo

Original Article

Factors independently associated with increased risk of pain development after ophthalmic surgery

Henzler, D.*†; Kramer, R.*; Steinhorst, U. H.; Piepenbrock, S.*; Rossaint, R.; Kuhlen, R.

Author Information
European Journal of Anaesthesiology: February 2004 - Volume 21 - Issue 2 - p 101-106


It is occasionally assumed that ophthalmic surgery tends to create little if any postoperative pain, which might be attributed to the limited surgical trauma compared to other disciplines. To enable rapid recovery from anaesthesia, short-acting anaesthetic agents are preferred. This might contribute to the development of early postoperative pain. Multiple procedures, special techniques or certain operation sites have been suspected to increase postoperative pain, but have not been investigated systematically. Preliminary reports have shown some pre-emptive analgesic effects in patients undergoing pars plana vitrectomy, if general anaesthesia was combined with regional anaesthesia [1,2]. However, little has been documented about the incidence and severity of postoperative pain after different ophthalmic surgical procedures. To assess the effect of any anaesthetic, analgesic or even pre-emptive analgesic regimens on the development of postoperative pain perception it is crucial to identify those factors, which are independently associated with the incidence and severity of pain development in these patients. Sufficient pain management is very important not only for patient comfort, but also to reduce the risk of postoperative cardiac and pulmonary complications [3-5]. Obviously, effective treatment and even more prevention of postoperative pain is a major objective of modern anaesthesia. Therefore, this observational study was planned to assess the incidence and severity of postoperative pain following ophthalmic surgery, and to identify key factors independently associated with development of postoperative pain.


Patient population

This prospective cohort study included 500 consecutive patients, who had undergone elective ophthalmic surgery and who were assessed for pain after operation. Written informed consent by the patients was obtained before operation. Children under the age of 18 yr and patients incapable of communicating adequately were excluded.

Surgical and anaesthetic procedures

Operations were classified into seven groups of equal anatomical location and with comparable surgical trauma each (Table 1). According to the usual clinical practice in our institution, general anaesthesia was performed on patients undergoing posterior segment surgery, long lasting or more extensive surgery and if regional anaesthesia was refused. Induction with propofol and alfentanil was followed by atracurium for muscle relaxation and orotracheal intubation. Anaesthesia was maintained with isoflurane 0.7-1.1% and nitrous oxide 60% in oxygen. Regional anaesthesia was performed by the surgeon, using bupivacaine 0.5% 2 mL and mepivacaine 3% 2 mL for retrobulbar block, or oxybuprocaine 0.7% for topical anaesthesia. The decision for either anaesthetic technique was based on the clinical condition of the particular patient and was agreed on between the surgeon and anaesthetist independently from the study.

Table 1
Table 1:
Classification of operations according to anatomical site and type of anaesthesia.

Study protocol

After surgery, pain scores and analgesic medication were assessed in each patient. Since many of the patients suffered significant visual impairment, a numerical analogue scale was used instead of a visual analogue scale (VAS) [6,7]. Endpoints were 0 for no pain and 10 for the most severe and unbearable pain. The numerical analogue scale is considered to be sufficiently sensitive, reliable and valid by various authors [8-11]. Assessments were made every 30 min for 2 h postoperatively, and in the following period up to 9 h after operation at 60 min intervals (t30, t60, t90, t2, t3,...). The last score was obtained on the following morning (t24). Patients were unrestricted in any analgesic medication required. Usual medication consisted of acetaminophen (paracetamol) 1 g orally, metamizol 1 g orally or intravenously (i.v.), or piritramide 3.75-7.5 mg i.v.


Analysis was performed by comparing the means of numerical analogue scales in each group, the incidence of serious pain (classified as numerical analogue scale ≥ 5) and postoperative analgesic medication required. Factors evaluated were the type and duration of surgery, age, gender, numbers of preceding operations and type of anaesthesia on level of pain development postoperatively. Mean numerical analogue scale ± SD scores within and between groups were compared using ANOVA and t-test with Bonferroni's test for multiple measurements. Categorical data were analysed using the χ2-test. Risk ratios (RR, 95% CI) were calculated for the suspected key factors for the risk of having a numerical analogue scale ≥ 5 any time postoperatively. The level of significance was set at P < 0.05 (all calculations performed with SPSS® Win 7.5 (SPSS Inc, Chicago, Ill, USA)).


Patients' characteristics are listed in Table 2. General anaesthesia was administered to 269 patients, while 231 patients received regional anaesthesia. On average, the latter were 9.6 yr older than patients receiving general anaesthesia (P < 0.05). The mean duration of surgery performed under regional anaesthesia was 29.2 ± 18.7 min compared to 62.7 ± 37.5 min, when performed under general anaesthesia (P < 0.05). The frequency of the various surgical procedures is shown in Table 1.

Table 2
Table 2:
Patients' characteristics.

Pain scores

The mean numerical analogue scale scores with time are shown in Figure 1, presenting maximum values during the period 90 min to 3 h postoperatively. Multivariate analysis demonstrated in Groups 2, 3, 4, 6 statistically significant higher scores than in Groups 1, 5, 7, which were present until 7 h postoperatively, and between Groups 1 and 2 at all times. Thus, posterior segment surgery, corneal surgery, ocular muscle surgery and enucleations were more likely to develop significant postoperative pain. A mean of 51.5% of these patients developed numerical analogue scale scores ≥ 5 at any time postoperatively, compared to only 19% of patients in Groups 1, 5, 7 (RR 4.5, CI 3.01-6.79, P < 0.0001). The individual risks of developing serious pain compared to the remainder of all patients examined are presented in Table 1. There was no significant difference between patients undergoing pars plana vitrectomy including encircling buckle (n = 50) vs. simple vitrectomy alone (n = 62).

Figure 1
Figure 1:
Mean numerical analogue scale after operation in accordance to procedure classification (Group 3 not charted because of low numbers). ―◆―: Group 2; ―▲―: Group 4; ―●―: Group 6; ―□―: Group 1; ―○―: Group 5; ―◇―: Group 7.

Key factors for pain development

Certain operations could be identified to result in significantly more pain after operation (Groups 2, 3, 4, 6, n = 169). Only these were used for further analysis, whereas the other groups with only little postoperative pain development were not considered for analysis of key factors. However, no factors were shown to be independently associated with an increased likelihood of a patient experiencing postoperative pain. The incidence of preceding ophthalmic operations (RR 0.93, CI 0.70-1.25), duration of surgery, (discriminative value: 60 min) (RR 1.02, CI 0.76-1.38), age (discriminative value: 70 yr, RR 0.84, CI 0.62-1.14), or preoperative co-morbidity according to ASA (III/IV vs. I/II: RR 1.11, CI 0.83-1.49) or particular disease showed no statistically significant impact on pain development characterized by a numerical analogue scales ≥ 5 at any time postoperatively. There was some difference as female patients experienced more pain (RR 1.58, CI 1.7-2.13, P = 0.002), but this did not result in more analgesic requirements in female patients (P = 0.275).

Influence of type of anaesthesia

Only in Group 1 were there comparable numbers of patients operated on under general anaesthesia (n = 86) and regional anaesthesia (n = 102). At all times from 60 min until 6 h postoperatively, and on the first postoperative day, patients receiving general anaesthesia had significantly higher numerical analogue scale scores than those operated on under regional anaesthesia (P < 0.001) (Fig. 2). The RR of developing serious pain after operation in general vs. regional anaesthesia was 6.52 (CI 2.33-18.2; P < 0.0001).

Figure 2
Figure 2:
Cataract operations: mean numerical analogue scale of operations performed under general (GA) and regional (RA) anaesthesia. ★:P < 0.001; ○: General anaesthesia; ▪: Regional anaesthesia.

Analgesic doses

Of 269 patients operated on under general anaesthesia, 116 patients (43%) requested analgesics, whereas only 48 of 231 patients (21%) operated on under regional anaesthesia requested analgesics after operation. There was a strong correlation between patients experiencing pain and given analgesics (RR 4.36, CI 3.33-5.71). As most of these patients were from Groups 2, 3, 4, 6, significantly more analgesic medication was given in these groups as compared to Groups 1, 5, 7 (RR 2.57, CI 1.98-3.33, P < 0.0001) (Fig. 3). No other factors could be shown to have any impact on analgesic medication.

Figure 3
Figure 3:
Percentage of patients receiving any analgesic dosage at postoperative observation times. (□: Groups 2, 3, 4, 6),n = 169: more severe pain development; (▪: Groups 1, 5, 7), n = 331: less severe pain development.

Pain characteristics

Wound sore was the most frequent pain and accounted for approximately 50%. The distribution of burning pain, blunt pain and headache was equal between pain characteristics as well as within time course.


The objective of this observational study was to assess the incidence and severity of postoperative pain following ophthalmic surgery and to identify key factors independently associated with development of postoperative pain. The occasionally held view that ophthalmic surgery is unlikely to cause serious postoperative pain may be related to the limited trauma compared to other surgical disciplines and thus is not immediately obvious to an external observer. This may account for the lack of published studies on this topic. We set out to investigate pain development after ophthalmic surgery systematically. Although an open, observational study is less favourable than a controlled trial, there was no valid data available as to which randomization could be performed. In recently published studies [1,2], - showing the pre-emptive analgesic effects of combining regional anaesthesia with general anaesthesia - just vitrectomy was chosen, not regarding the fact that other kinds of operation might create more pain. Also, if additional regional anaesthesia for pain control is administered the risk of postoperative pain developing in any individual patient should be well known, because it imposes an additional, though very low, risk of severe complications [12-15]. Nothing is known about possible co-factors associated with increased postoperative pain. Our aim was to gather valid data prospectively on the development of pain after ophthalmic surgery. We used a numerical analogue scale scoring system, which has been previously shown to be as reliable, sensitive and valid, as the well-established VAS [9]. Surprisingly, we found out only two conditions under which the likelihood of developing serious pain following ophthalmic surgery was increased. These are described as follows:

Operation site. Posterior segment surgery, surgery to the cornea and muscles, and enucleation, were more likely to develop higher numerical analogue scale scores than anterior segment surgery, e.g. cataract extraction, and surgery to the lid apparatus. The classification into the seven groups was arbitrary, but was defined according to similar operational trauma in the different anatomical regions of the eye. This can be justified, since no published classification scheme exists and pain distribution within groups was quite homogenous. Individual multivariate analysis indeed exhibited a strong correlation between these groupings and the development of pain. We do consider enucleation as creating more pain, even though the risk ratio did not show significance (P = 0.352). Still the percentage of patients in this group with serious pain is considerably higher than the average, and the small number of patients might have prevented the difference from becoming significant. The fact that as many as 52% of the patients in Groups 2, 3, 4, 6 experienced serious pain, with three of them suffering unbearable pain (numerical analogue scale = 10), is not only a remarkable result for operations often prejudiced to be 'pain-free', but also sheds some light on deficits in postoperative pain management.

Type of anaesthesia. There were quite different surgical procedures performed in this open, observational study, which makes it impossible to compare general anaesthesia and regional anaesthesia between different groups. Consequently, regarding this issue, we only analysed the patients in Group 1, with nearly equal numbers receiving general anaesthesia and regional anaesthesia. When bupivacaine was the local anaesthetic agent used for regional anaesthesia, in most of these patients the numerical analogue scale scores were reduced for 6 h after surgery, a time period likely to be covered by the regional anaesthesia. Patients who received general anaesthesia had significantly higher numerical analogue scale scores. This study was not designed to compare general anaesthesia with regional anaesthesia, hence the mean pain scores in this subgroup were low, yet significantly different. If translated for use in more painful operations the beneficial effect of regional anaesthesia, even if administered solely for postoperative analgesia, should be clear and has been demonstrated in controlled studies [1,2]. Even though most ophthalmic operations can be carried out under regional anaesthesia, patient disagreement, requirement to avoid any patient movement, or inability of the patient to lie supine are indications for general anaesthesia. Most pars plana vitrectomies were performed for retinal detachment and subretinal surgery and are routinely performed under general anaesthesia in our institution, often performed on 'the last eye' with operating times up to a maximum of 390 min. This, and a patient population typical of a university hospital setting, may account for the high number of operations performed under general anaesthesia (54%) [16].

The following factors were not shown to influence postoperative pain development: age, ASA classification of disease and revision surgery. This should totally eliminate the sometimes-expressed prejudice that there is less postoperative pain in elderly or sick people. Female patients expressed higher numerical analogue scale scores, but did not take more analgesics than male patients. The difference, even though significant, was small and this study was not designed to answer the question whether the observed difference is attributable to real differences in pain or to gender specific differences in tolerance and attitude towards pain. The assumption that pars plana vitrectomy including encircling buckle would develop higher numerical analogue scale scores than plain vitrectomy alone, was not supported by the results of this study.

Our data suggests that improvement and extension of postoperative analgesia can be achieved by the use of long acting local anaesthetics (e.g. ropivacaine) or by a sub-Tenon's block given by the surgeon at the end of a posterior segment procedure. Studies investigating pre-emptive or preventive analgesia should focus on patients undergoing surgery of the posterior segment, eye muscles or cornea, because only in these ophthalmic patients considerable pain justifying an additional regional anaesthesia is experienced.


Since serious pain was not infrequently found despite pain monitoring, the need for anaesthetic techniques that reduce or even prevent postoperative pain is obvious. Adequate pain management, ideally resulting in the complete absence of postoperative pain, not only provides comfort to such patients, but may also contribute to improved healing and a reduction in the incidence of postoperative complications [16,17].

No factors independently associated with increased postoperative pain could be identified. Patients undergoing surgery of the posterior segment, eye muscles or cornea were more likely to develop serious pain after surgery. This was specifically relevant, since these operations were carried out under general anaesthesia, which in a subgroup analysis showed to be more likely to create postoperative pain than if performed under regional anaesthesia.


1. Kristin N, Schonfeld CL, Bechmann M, et al. Vitreoretinal surgery: pre-emptive analgesia. Br J Ophthalmol 2001; 85: 1328-1331.
2. Henzler D, Müller-Kaulen B, Steinhorst UH, Broermann H, Piepenbrock S. The combination of retrobulbar block with general anaesthesia may lead to pre-emptive analgesia in patients undergoing pars plana vitrectomy. Anaesthesiol Intensivmed Notfallmed Schmerzther 2002; 37: 267-272.
3. Rawal N, Sjöstrand U, Christoffersson E, Dahlstrom B, Arvill A, Rydman H. Comparison of intramuscular and epidural morphine for postoperative analgesia in the grossly obese: influence on postoperative ambulation and pulmonary function. Anesth Analg 1984; 63: 583-592.
4. Scott NB, Kehlet H. Regional anaesthesia and surgical morbidity. Br J Surg 1988; 75: 299-304.
5. Tuman KJ, McCarthy RJ, March RJ, et al. Effects of epidural anesthesia and analgesia on coagulation and outcome after major vascular surgery. Anesth Analg 1991; 73: 696-794.
6. Duncan GH, Bushnell MC, Lavigne, GJ. Comparison of verbal and visual analogue scales for measuring the intensity and unpleasantness of experimental pain. Pain 1989; 37: 295-303.
7. Yarnitsky D, Sprecher E, Zaslansky R, Hemli JA. Multiple session experimental pain measurement. Pain 1996; 67: 327-333.
8. Downie WW, Leatham PA, Rhind VM, Wright V, Branco JA, Anderson JA. Studies with pain rating scales. Ann Rheum Dis 1978; 37: 378-381.
9. Ferraz MB, Quaresma MR, Aquino LRL, Atra E, Tugwell P, Goldsmith CH. Reliability of scales in the assessment of literate and illiterate patients with rheumatoid arthritis. J Rheumatol 1990; 17: 1022-1024.
10. Jensen MP, McFarland CA. Increasing the reliability and validity of pain intensity measurement in chronic pain patients. Pain 1993; 55: 195-203.
11. Price DD, Bush FM, Long S, Harkins SW. A comparison of pain measurement characteristics of mechanical visual analogue and simple numerical rating scales. Pain 1994; 56: 217-226.
12. Feitl ME. Anaesthesia. In: Krupin T, Kolker AE, eds. Complications of Ophthalmic Surgery. Baltimore, USA: Wolfe, 1993: 1-18.
13. Hamilton RC. Brain stem anesthesia following retrobulbar blockade. Anesthesiology 1985; 63: 688-690.
14. Rosenblatt RM, May DR, Barsonmian K. Cardiopulmonary arrest after retrobulbar block. Am J Ophthalmol 1980; 90: 425-427.
15. Wittpenn JR, Rapoza P, Sternberg P, Kuwashima L, Saklad J, Patz A. Respiratory arrest following retrobulbar anesthesia. Ophthalmology 1986; 93: 867-870.
16. Seitz W, Schäffer J. Perioperativer Stress bei Augenoperationen. In: Piepenbrock S, Schäffer J, eds. Anästhesie in der Augenheilkunde. Schriftenreihe Intensivmedizin, Notfallmedizin, Anästhesiologie. Stuttgart, Germany: Thieme, 1989: 45-54.
17. Woolf CJ, Chong MS. Preemptive analgesia - treating postoperative pain by preventing the establishment of central sensitization. Anesth Analg 1993; 77: 362-379.


© 2004 European Academy of Anaesthesiology