Haemorrhage and risk factors associated with retrobulbar/peribulbar block: a prospective study in 1383 patients

H. Kallio, M. Paloheimo and E.-L. Maunuksela

Department of Anaesthesia, Helsinki University Central Hospital, Helsinki University Eye Hospital, PO Box 220, FIN-00029 Helsinki, Finland

Accepted for publication: 29 June, 2000


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients undergoing intraocular surgery are elderly and may have disease or be receiving medication which increases the risk of haemorrhage. We interviewed 1383 consecutive patients scheduled for eye surgery requiring retrobulbar/peribulbar block about their use of non- steroidal anti-inflammatory drugs, oral steroids and warfarin. A history of diabetes mellitus and globe axial length was noted. Medial peribulbar and inferolateral retrobulbar blocks were performed by three specialists and six doctors in training. The ensuing haemorrhages were graded as follows: 1=spot ecchymosis; 2=lid ecchymosis involving half of the lid surface area or less; 3=lid ecchymosis all around the eye, no increase in intraocular pressure; 4=retrobulbar haemorrhage with increased intraocular pressure. Acetylsalicylic acid was taken by 482 (35%) patients, non-steroidal anti-inflammatory drugs by 260 (19%) and warfarin by 76 (5.5%). Lid haemorrhages (grades 1–3) were observed in 55 patients (4.0%); in 33 of these patients the haemorrhages were spotlike (grade 1). No grade 4 haemorrhages occurred. The preoperative use of acetylsalicylic acid, non-steroidal anti-inflammatory drugs or warfarin, whether or not they had been discontinued, did not predispose to haemorrhage associated with retrobulbar/peribulbar block.

Br J Anaesth 2000; 85: 708–11

Keywords: surgery, ophthalmological; complications, haemorrhage; anaesthesia, local; non-steroidal anti-inflammatory drugs


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Patients scheduled for intraocular surgery are often elderly and may have concurrent diseases. In the Finnish population aged over 65 yr, non-steroidal anti-inflammatory drugs (NSAIDs) are prescribed to 27.9% of people, mostly for degenerative musculoskeletal diseases (Prescription Register, 1996, Social Insurance Institution, Finland). In this age-group, warfarin is used for anticoagulation by 5.0% of people in order to prevent thromboembolism, and a low dose of acetylsalicylic acid is used by 4.9% to prevent myocardial infarction and cerebrovascular ischaemic events (Prescription Register, 1996, Social Insurance Institution, Finland). Haemorrhage is a rare complication of regional ophthalmic anaesthesia: lid ecchymoses (2.3–3.5%)1 and retrobulbar haemorrhage (0.44–1.7%)2 3 4 are the most common. As the benefits of retrobulbar/peribulbar block may outweigh the risk of puncture-induced haemorrhagic complications, even in patients on anticoagulant medication, we undertook a prospective study of the relationship between the use of anticoagulant drugs and bleeding associated with retrobulbar/peribulbar block. We assessed the influences of known risk factors related to the patients (e.g. anticoagulant medication, extended axial length of the eye) and the experience of the physician performing the block on the incidence of bleeding.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Approval of the hospital ethics committee was obtained for the study. During 4 months of 1997 at Helsinki University Eye Hospital, 1383 consecutive patients scheduled for eye surgery or laser treatment under retrobulbar/peribulbar block were interviewed about their use of acetylsalicylic acid, NSAIDs, paracetamol, warfarin and oral steroids and their self-estimated tendency to develop haematomas. On the morning of operation, thromboplastin time (Stago Prothrombin Complex Assay, Diagnostica Stago, Asnieres, France) was measured from blood samples of the patients taking warfarin. In our hospital, patients are advised to discontinue warfarin for 2 days and acetylsalicylic acid for 3 days before surgery. Non-compliance with these rules does not usually lead to cancellation of surgery. Data on ophthalmic diagnosis and surgical procedure, existing diabetes mellitus, globe axis length, sex and age were recorded. It is considered difficult to perform retrobulbar injection if the globe axial length exceeds 26 mm,5 and in such cases it is more likely that repeated attempts will be needed to perform the injection than in patients with normal anatomy.

The appearance and size of the haemorrhages were assessed on a four-grade scale: 1=spot ecchymosis; 2=lid ecchymosis involving half the lid surface area or less; 3=lid ecchymosis all round the eye, no increase in intraocular pressure; 4=retrobulbar haemorrhage with increased intraocular pressure. Nine doctors performed the blocks; three were experienced anaesthetists and six were doctors in training (four anaesthetists and two ophthalmologists).

Transconjunctival retrobulbar/peribulbar block was performed as described previously.6 First, a medial peribulbar puncture was made with a sharp, disposable, 12-mm, 30-gauge needle (Microlance®; Becton Dickinson, Drogheda, Ireland) in the near perpendicular direction of the frontal plane, followed by inferolateral retrobulbar insertion of a sharp, disposable, 31-mm, 27-gauge needle (PrecisionGlide®; Becton Dickinson, Franklin Lakes, NJ, USA). The local anaesthetic used was a 1:1 mixture of 0.75% bupivacaine and 2% lidocaine with hyaluronidase 0–7.5 IU ml–1. The maximum total volume of local anaesthetic was adjusted according to the lean body weight of the patient: 6 ml for patients under 70 kg body weight, 7 ml for patients between 70 and 80 kg, and 8 ml for those over 80 kg. As the local anaesthetic was injected, the anaesthetist palpated the lid with his finger, and if increased pressure was sensed the injection was stopped. Orbital compression of 25 mm Hg for 10 min was accomplished with an autopressor (Autopressor®; Storz, Heidelberg, Germany), except in patients who had had an operation on the same eye during the last 6 months. All patients were monitored using finger-pulse oximetry, continuous end-expiratory carbon dioxide using tubing in the nostril, electrocardiography, and non-invasive automatic blood pressure every 10 min. An intravenous cannula was inserted for the administration of sedatives or fentanyl as required. A stream of oxygen-enriched air (oxygen 2 litre min–1 + air 8 litre min–1) was directed at the patient’s face under the surgical drape.

The hypothesis was that use of anticoagulants and the preoperative discontinuation time do not affect the occurrence of haemorrhages (grades 1–4). Statistical analysis was undertaken with the Mann–Whitney rank sum, {chi}2 and Fisher’s exact tests, using SigmaStat for Windows, version 2.0 (Jandel Corporation). Values are expressed as mean (SD, range). A P value <0.05 was considered statistically significant.


    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The physical characteristics and the ophthalmic diagnoses did not differ between patients with and without haemorrhage (Table 1). Senile nuclear cataract was the most common diagnosis (1061 patients (77%)), and most of the patients with this diagnosis (1029) underwent extracapsular operation by a phacoemulsification technique. The operation on one cataract patient was cancelled because of persisting nausea immediately after the block. One hundred and seventeen patients (7.8%) were excluded from the study. Seven patients gave insufficient history because of senile dementia, 27 had missing documentation about the haemorrhages, and 83 patients had other data missing.


View this table:
[in this window]
[in a new window]
 
Table 1 Physical characteristics (mean (SD, range)) and ophthalmic diagnoses in patients with and without haemorrhage
 
The overall frequency of haemorrhages was 4.0% (55 of 1383 patients). The haemorrhages were distributed into the following grades: grade 1, 33 patients; grade 2, 15 patients; grade 3, 7 patients. No high-pressure retrobulbar haemorrhages (grade 4) occurred in this series. Only one scheduled operation was cancelled because of haemorrhage (grade 3), although the intraocular pressure was only 13 mm Hg and no decompressive operative measures were needed. It was noted that this patient sneezed during the peribulbar injection. In the preoperative interview, a haemorrhagic tendency was reported by 32% of patients but it did not predispose to haemorrhage. Trainees did not evoke significantly more haemorrhages (4.7%) than staff anaesthetists (3.8%).

Use of more than one drug that may affect coagulation was characteristic of this patient population (Table 2). Also, the drug-free preoperative period varied considerably. The drug category, the time of discontinuing drug therapy and the use of various combinations of drugs did not affect significantly the occurrence of haemorrhage. Warfarin was used by 5.5% of the patients and its use did not predispose to haemorrhage (Table 2). The mean thromboplastin time value in patients who had haemorrhage (35% (19, 20–56)) did not differ significantly from that of patients who had no haemorrhage: (39% (16, 14–81)). The prevalence of diabetes mellitus was 22% in patients receiving warfarin, and cortisone was used continuously by 3.9% of diabetics receiving warfarin. Among the latter subgroup (three patients), none developed any detectable haemorrhage from the block. Table 2 shows the frequencies of concomitant use of warfarin with acetylsalicylic acid, NSAIDs or paracetamol.


View this table:
[in this window]
[in a new window]
 
Table 2 Preoperative anticoagulants and haemorrhage
 
Oral steroids were used by 4.2% of the patients but did not predispose to haemorrhage (1.7%). Patients on cortisone therapy, however, reported a haemorrhagic tendency significantly more often than those receiving no steroids (56 and 32% respectively (P<0.001)). Cortisone use was significantly more frequent among diabetics than non-diabetics (7.5 and 3.6% respectively (P=0.021)). The prevalence of diabetes mellitus in the whole sample was 15%, and the incidence of haemorrhage in the diabetic group was 4.5%.


    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The overall incidence of lid haemorrhages in our study (4.0%) was similar to the incidences in earlier reports (2.3–3.5%).1 Preoperative use of acetylsalicylic acid or other NSAIDs did not increase the frequency and grade of haemorrhages associated with retrobulbar/peribulbar block. The interval from discontinuation of these drugs until surgery did not affect haemorrhage.

It can be assumed that visible haemorrhages do not always appear within 10 min (duration of autopressor compression) after injecting the block and a few cases may have remained undetected, especially those presenting as minor spot-like haemorrhages. The physician performing the retrobulbar/peribulbar block graded the occurrence and size of the haemorrhage. We are aware that a non-blinded assessment could have been biased and, to some extent, led to underestimation of iatrogenic complications. But the observed frequency of haemorrhages, as evaluated 10 min after the injection, is probably accurate. It is also possible that patients who had taken acetylsalicylic acid and/or NSAIDs 1–2 weeks before surgery were registered as non-users. Elderly patients may not remember occasional use of these drugs. Such assumed inaccuracy of the data may make the results somewhat less powerful. Haemorrhages were as frequent in patients taking warfarin as in those who were not. More than 500 patients receiving warfarin should have been included in this study in order to make any conclusion about the necessity of stopping warfarin preoperatively. Furthermore, if more patients had been included it may have been possible to detect any interaction of warfarin and platelet aggregation inhibitors in provoking haemorrhage.

A large retrobulbar haemorrhage may delay a planned eye operation,2 and may cause compression of the circulation7 or optic atrophy8 leading to permanent blindness. The prevalence of retrobulbar haemorrhage after retrobulbar injection has been reported to vary from 0.44 to 1.7%.24 In a retrospective study 60 retrobulbar haemorrhages were induced by a retrobulbar block (1.7% of all retrobulbar blocks). Warfarin, acetylsalicylic acid and NSAIDs were stopped 14 days before surgery. The study included a control group of 60 patients. Thirty per cent of patients with haemorrhage received at least one of these drugs, whereas these drugs were taken by only 20% of the patients in the control group.4

Platelet cyclo-oxygenase is acetylated irreversibly by acetylsalicylic acid. Cyclo-oxygenase inhibition by NSAIDs leads to reversible disturbance of platelet aggregation.9 A healthy individual produces platelets at the rate of about 70x109 litre–1 per day and a concentration of normal functioning platelets of 30–50x109 litre–1 is adequate for normal clotting.10 Forty-eight hours should, therefore, be sufficient for the recovery of clinically adequate thrombocyte function after the last mini-dose of acetylsalicylic acid.10 There have been no large-scale studies of the influence of the duration of discontinuation of acetylsalicylic acid, other NSAIDs and warfarin on haemorrhage during retrobulbar/peribulbar blocks. But there are some similarities between retrobulbar/peribulbar and epidural blocks. Whereas the use of warfarin at therapeutic doses contraindicates epidural block, the use of acetylsalicylic acid and other NSAIDs up to the day of the block may be considered safe in patients undergoing epidural block.11 12 Acetylsalicylic acid 60 mg day–1 is not associated with a significant increase in bleeding during epidural anaesthesia.11 The use of antiplatelet therapy in patients undergoing spinal or epidural anaesthesia has not been found to be associated with an increased incidence of spinal haematoma.12 However, acetylsalicylic acid medication has been reported to increase bleeding at the operation site.1315 In contrast, there is some disagreement about the role of other NSAIDs in surgical bleeding, but on the whole it does not increase it.1619 The exception is a report of 41 hysterectomy patients who received indomethacin preoperatively for postoperative analgesia.20

As a result of complications associated with retrobulbar block, several alternative techniques have been developed, such as the use of a curved needle. This was associated with a very low frequency of haemorrhages (lid ecchymosis 2%).21 Topical anaesthesia is also free from anaesthesia-related haemorrhage. However, the sub-Tenon method of regional ophthalmic anaesthesia may not be an option for reducing the risk of haemorrhage when compared with retrobulbar/peribulbar anaesthesia.22

In our practice there seems to be no reason to limit the use of acetylsalicylic acid or other NSAIDs before retrobulbar/peribulbar block. On the other hand, our study supports the assumption that preoperative discontinuation of warfarin for 2 days preoperatively might be sufficient to maintain haemostasis to prevent bleeding during retrobulbar/peribulbar block.


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1 Hamilton RC, Gimbel HV, Strunin L. Regional anaesthesia for 12,000 cataract extraction and intraocular lens implantation procedures. Can J Anaesth 1988; 35: 615–23[Abstract]

2 Edge KR, Nicoll JMV. Retrobulbar hemorrhage after 12,500 retrobulbar blocks. Anesth Analg 1993; 76: 1019–22[Abstract]

3 Ruben S. The incidence of complications associated with retrobulbar injection of anaesthetic for ophthalmic surgery. Acta Ophthalmol 1992; 70: 836–8

4 Cionni RJ, Osher RH. Retrobulbar hemorrhage. Ophthalmology 1991; 98: 1153–5[ISI][Medline]

5 Rubin AP. Complications of local anaesthesia for ophthalmic surgery. Br J Anaesth 1995; 75: 93–6[Free Full Text]

6 Sarvela PJ, Paloheimo MPJ, Nikki PH. Comparison of pH-adjusted bupivacaine 0.75% and a mixture of bupivacaine 0.75% and lidocaine 2%, both with hyaluronidase, in day-case cataract surgery under regional anesthesia. Anesth Analg 1994; 79: 35–9[Abstract]

7 Sullivan KL, Brown GC, Forman AR, Sergott RC, Flanagan JC. Retrobulbar anesthesia and retinal vascular obstruction. Ophthalmology 1983; 90: 373–7 [ISI][Medline]

8 Ellis PP. Retrobulbar injections. Surv Ophthalmol 1974; 18: 425–30

9 Cronberg S, Wallmark E, Söderberg I. Effect on platelet aggregation of oral administration of 10 non-steroidal analgesics to humans. Scand J Haematol 1984; 33: 155–9[ISI][Medline]

10 Roizen MF. Anesthetic implications of concurrent diseases. In: Miller RD, ed. Anesthesia. Philadelphia: Churchill Livingstone, 2000; 903–1015

11 CLASP (Collaborative Low-dose Aspirin Study in Pregnancy) collaborative group. CLASP: a randomised trial of low-dose aspirin for the prevention and treatment of pre-eclampsia among 9364 pregnant women. Lancet 1994; 343: 619–29[ISI][Medline]

12 Horlocker TT, Wedel DJ, Schroeder DR, et al. Preoperative antiplatelet therapy does not increase the risk of spinal hematoma associated with regional anesthesia. Anesth Analg 1995; 80: 303–9[Abstract]

13 Stage J, Jensen JH, Bonding P. Post-tonsillectomy haemorrhage and analgesics. A comparative study of acetylsalicylic acid and paracetamol. Clin Otolaryngol 1988; 13: 201–4[ISI][Medline]

14 Watson CJE, Deane AM, Doyle PT, Bullock KN. Identifiable factors in post-prostatectomy haemorrhage: the role of aspirin. Br J Urol 1990; 66: 85–7[ISI][Medline]

15 Thurston AV, Briant SL. Aspirin and post-prostatectomy haemorrhage. Br J Urol 1993; 71: 574–6[ISI][Medline]

16 Bricker SRW, Savage ME, Hanning CD. Peri-operative blood loss and non-steroidal anti-inflammatory drugs: an investigation using diclofenac in patients undergoing transurethral resection of prostate. Eur J Anaesthesiol 1987; 4: 429–34[ISI][Medline]

17 Power I, Noble DW, Douglas E, Spence AA. Comparison of i.m. ketorolac trometamol and morphine sulphate for pain relief after cholecystectomy. Br J Anaesth 1990; 65: 448–55[Abstract]

18 Rømsing J, Østergaard D, Walther-Larsen S, Valentin N. Analgesic efficacy and safety of preoperative versus postoperative ketorolac in paediatric tonsillectomy. Acta Anaesthesiol Scand 1998; 42: 770–5[ISI][Medline]

19 Tarkkila P, Saarnivaara L. Ketoprofen, diclofenac or ketorolac for pain after tonsillectomy in adults? Br J Anaesth 1999; 82: 56–60[Abstract/Free Full Text]

20 Engel G, Lund B, Kristenson SS, Axel C, Nielsen JB. Indomethacin as an analgesic after hysterectomy. Acta Anaesthesiol Scand 1989; 33: 498–501[ISI][Medline]

21 Teichmann KD, Uthoff D. Retrobulbar (intraconal) anesthesia with curved needle: technique and results. J Cataract Refract Surg 1994; 20: 54–60[ISI][Medline]

22 Hamilton RC. Complications of ophthalmic regional anesthesia. In: Finucane BT, ed. Complications of Regional Anesthesia. Philadelphia: Churchill Livingstone 1999; 39–55