Spinal anaesthesia is a simple, highly effective, well tolerated and economic technique. Despite various advantages of this form of analgesia, there are some complications. Recurrent, bilateral or unilateral, persistent or transient, mild or profound hearing loss, especially in the low-frequency region, has been reported after spinal anaesthesia [1–5]. The mechanism of transient hearing loss is attributed to leakage of cerebrospinal fluid (CSF), which leads to a decrease in perilymph pressure within the cochlea . It has been reported that the amount of fluid leakage, and thus the incidence of hearing loss and other possible complications such as headache, is related to the design and diameter of the spinal needle used [7–9]. Several studies have been published investigating the effects of differences in the designs of the needles on hearing loss after spinal anaesthesia, but we are not aware of any comparative study with the ballpen spinal needle in literature. The ballpen spinal needle was designed with a conical pencil-like tip formed by a stylet, which is withdrawn after penetration of the dura. After removal of the stylet, the spinal needle presents a relatively large hole at the tip of the needle that, theoretically, allows subsequent and unimpaired CSF flow . The aim of the present study was to determine the degree of hearing loss in patients undergoing spinal anaesthesia for inguinal herniorraphy. Three types of spinal needle were compared during a study of auditory function following spinal anaesthesia.
Participants and methods
After obtaining informed consent from patients and approval of the local ethics committee, 45 ASA physical status I patients scheduled for elective inguinal herniorraphy with spinal anaesthesia were enrolled in the study. The following were the exclusion criteria: patients with ASA II or above, a history of hearing disorders, a previous head trauma, chronic smoking and the inability to cooperate during audiometry. The patients were not premedicated. The patients were randomly divided into three groups. Group Q (n = 15) patients received spinal anaesthesia through a 25-gauge (G) Quincke spinal needle (Spinocan; B|Braun, Melsungen, Germany), group B (n = 15) patients received the same through a 25-G ballpen spinal needle (SpinoStar; Rusch, Betschdorf, France) and those in group P (n = 15) received the same through a 25-G pencil-point spinal needle (Pencan; B|Braun, Melsungen, Germany). They were monitored by SpO2, electrocardiogram and noninvasive blood pressure. Spinal anaesthesia was performed via the L3–4 or L4–5 interspace, with 15 mg hyberbaric bupivacaine intrathecally in all groups. Spinal anaesthesia was performed by the same anaesthesiologist with a midline approach and with the patient in the sitting position. Only one dural puncture was allowed in each patient; if an additional puncture was necessary, the patient was excluded from the study. Ten minutes after spinal injection, the maximum height of the block was assessed by pin-prick testing and was monitored at 5 min intervals for 30 min. Midazolam was used for sedation intravenously. All patients received 500 ml of intravenous (i.v.) 0.9% NaCl solution at 6 ml kg−1 h−1 before spinal anaesthesia. The infusion was continued at 4 ml kg−1 h−1, and a total of 3 l of balanced electrolyte solution (Isolyte-S; Eczacibasi-Baxter, Istanbul, Turkey) or 0.9% isotonic NaCl was administered over the 24 h perioperative period. Level of sensory block, heart rate, pulse oximeter values and noninvasive blood pressure were recorded every 5 min. If necessary, intravenous ephedrine was given to maintain systolic blood pressure (SBP) above 90 mmHg. During postoperative days 1 and 2, all patients were visited by an anaesthesiologist who was blinded to the type of spinal needle. Patients were interviewed about postoperative complaints such as postdural puncture headache (PDPH), vertigo, nausea–vomiting, transient neurological symptoms and major neurological deficits.
All patients were routinely examined by the same otolaryngologist preoperatively to assess middle-ear disease. Postoperatively, all patients were re-examined to exclude development of any acute middle-ear disease, which could lead to conductive hearing loss. Audiometric evaluation was made with pure-tone audiometry (AC 30 Clinical Audiometer; Interacoustic, Denmark) by the same audiologist at specific time intervals: preoperative, postoperative day 1 and postoperative day 2 in the noise-free audiometry laboratory. Hearing levels were tested at frequencies of 250–6000 Hz at these time intervals. To assess changes in hearing level at each frequency, the bone and air conduction hearing thresholds on the preoperative audiograms were compared with those on the postoperative day 1 and 2 audiograms. A change of more than 10 dB between preoperative and postoperative hearing levels was defined as a clinically significant change in hearing patterns. The mean change in the hearing level for both ears was calculated at each frequency. Also, the number of the cases that showed any change greater than 10 dB in hearing threshold at any frequency was determined in each group. Neither the patients nor the audiologist knew what type of needle was used for patients in the study groups.
Statistical analyses were performed by using SPSS for Windows (SPSS, Chicago, Illinois, USA). Comparison of variations in the right and left ear in hearing thresholds was performed by Mann–Whitney U-tests. Comparison of cases found to have hearing loss greater than 10 dB was carried out by the chi-squared test. Statistical analysis included the Friedman test, Wilcoxon signed-ranks test (with Bonferroni correction), Kruskal–Wallis and Mann–Whitney U-tests (with Bonferroni correction) for multiple comparisons. A P value less than 0.05 was considered significant.
There were no significant differences in patient data, operation times and maximal height of spinal block (P > 0.05; Table 1). No patient experienced vertigo, transient neurological symptoms or major neurological deficits after spinal anaesthesia. The preoperative and intraoperative blood pressure values in all groups were similar (P > 0.05). Preoperative, intraoperative and postoperative volume replacement was also comparable in all groups.
There were no significant differences between the left and right ear with respect to the decibel level at which the patient could hear any of the frequencies. Therefore, only those results from one ear (the right ear) are presented for all groups. If the patient required a higher decibel level (compared with the preoperative value) to hear a particular frequency, it signified hearing loss. Measurements of air and bone conduction hearing thresholds were equal at any frequency tested. No gap was found between air and bone conduction hearing thresholds. A change of at least 10 dB represents clinically significant changes in hearing patterns.
The mean preoperative, postoperative day 1 and postoperative day 2 hearing thresholds in each frequency are reported in Tables 2–4. When the three groups were compared for mean change in hearing between the preoperative level and that on postoperative day 1, a statistically significant difference was observed at 250, 500, 1000, 2000 and 4000 Hz (P < 0.05). At 6000 Hz frequency, the difference was insignificant (P > 0.05) (Table 5). When group B and group P were compared for change in hearing, no statistically significant difference was detected at any frequency tested. There was no difference between the groups with respect to the mean changes in hearing between the preoperative level and that on postoperative day 2 (P > 0.05). Mean changes in the hearing levels between preoperative and postoperative day 1/postoperative day 2 values and comparison of the three groups in terms of the mean change in the hearing level are reported in Table 5. There were statistically significant differences between preoperative and postoperative day 1 measurements at 250, 500, 2000 and 4000 Hz frequencies between group Q and group B/P. No difference was found between group Q and B in terms of mean preoperative and postoperative day 1 changes at 1000 Hz frequency. At this frequency, there was a significant difference between groups Q and P (Table 5). There were significant differences between the three groups in terms of the number of patients who had greater than 10 dB hearing loss at 250, 500, 4000 and 6000 Hz on postoperative day 1. The number of patients who had greater than 10 dB hearing loss in group Q was significantly more than that found in group B and group P. Table 6 gives an overview of the details of the number of patients according to the groups.
In all patients, the amount of intraoperative bleeding was negligible. No patient complained of hearing loss. Only one patient had nausea–vomiting in group Q. There was no difference between the groups with respect to nausea–vomiting (P > 0.05). Two patients who had greater than 10 dB hearing loss from group Q complained of PDPH on postoperative day 1. They were medicated with i.v. fluids and bed rest.
We studied the effects of the needle type on postspinal anaesthesia hearing loss with the use of pure-tone audiometry. At 250, 500, 2000 and 4000 Hz, group Q showed significant hearing loss between the preoperative level and that on postoperative day 1 compared with group B and group P. When group B and group P were compared for changes in hearing, no statistically significant difference was detected at any frequency tested. The number of patients who had greater than 10 dB hearing loss in group Q was significantly more than that found in group B or group P at 250, 500, 4000 and 6000 Hz on postoperative day 1.
Spinal anaesthesia is one of the most frequently used regional anaesthesia techniques in surgical interventions; however, rarely it may cause some neurological problems. One of these problems is hearing loss [1,7,11,12]. Although Wang et al. reported hearing loss lasting as long as 7 months after spinal anaesthesia, Kilickan et al. reported a case of hearing loss after spinal anaesthesia lasting more than 2 years. PDPH, vertigo and tinnitus may be clinical implications of hearing loss after spinal anaesthesia . Two patients from group Q complained of PDPH in this study. No patient had vertigo or tinnitus in all groups in the present study. Audiometry may be a more sensitive indicator of CSF leak than PDPH . Although hearing loss after spinal anaesthesia has been described, it is not generally considered a common complication of this technique, perhaps because most patients do not notice or report hearing loss . We also observed that no patient complained of hearing loss in the present study.
Several studies have been published investigating the incidence of hearing loss after spinal anaesthesia [2,3,7,8,11,14–16]. The incidence of hearing loss following spinal anaesthesia depends on several parameters and has shown great variability in previous reports.
The anatomy of hearing is complex but can be divided into the peripheral part (external, middle and inner ear, and the cochlear and vestibular divisions of the auditory nerve) and the central part (hearing pathways, subcortical and cortical auditory centres and central balance mechanism). The CSF dynamics are important for auditory function of the inner ear. The membranous labyrinth encloses endolymph that fills a hollow system within the inner ear. Perilymph, a filtrate of blood and CSF, is the substrate for the inner ear cells and is present in the cochlea. There is passive diffusion as well as active transport of ions between the endolymph and the perilymph. The mechanism for hearing loss after spinal anaesthesia is attributed to the disruption of this endolymph/perilymph balance caused by the decreased CSF pressure . After a leak via the spinal puncture hole, the perilymph passes into the subarachnoid space via the cochlear aqueduct because of the decrease in CSF pressure. Then, perilymph pressure decreases as well. In the meantime, endolymphatic and perilymphatic pressures change, and there seems to be a relative increase in endolymphatic pressure, resulting in the formation of an endolymphatic hydrops. An endolymphatic hydrops displaces the hair cells on the basement membrane and results in low-frequency hearing loss [7,18–20]. Walsted  also observed a parallel increase in the number of patients with hearing loss and the estimated loss of CSF after spinal anaesthesia, neurosurgical operations and acoustic neuroma resection.
That the amount of fluid leakage, and thus the incidence of hearing loss, is related to the diameter of spinal needle used has been reported. Fog et al.[7,8] reported a 92% incidence of decreased hearing level with the use of a 22-G spinal needle but only a 29% incidence with the use of a 26-G needle. This incidence ranged between 2 and 42% in other studies [3,5,8]. However, Finegold et al.  reported no hearing loss in 40 patients who underwent spinal anaesthesia for elective Caesarean delivery with different types of needle (24-G pencil-point and 25-G cutting tip). This discrepancy between reports is difficult to explain. In our study, 25-G (Quincke, ballpen, pencil-point) needles were used for all patients.
Wang et al. have suggested that, in addition to the diameter of needle, the irrigation solution used in transurethral resection is also a factor involved in hearing loss.
In-vitro studies have shown that dural leakage is related to the shape of the needle . Differences in the designs of the needles may affect the size of CSF leakage and thereby influence the CSF pressure. Sundberg et al. have suggested that one of the factors influencing the amount of CSF leak is the shape of the needle. In the previous study, 22-G Whitacre and Quincke needles were used, with more frequent hearing loss in cases of spinal anaesthesia with the 22-G Quincke needles . In the present study, the number of patients who had greater than 10 dB hearing loss in group Q was also significantly more than that found in group B and group P at 250, 500, 4000 and 6000 Hz on postoperative day 1. In a similar study, Malhotra et al. compared the effect of 22-G cutting-type and 25-G noncutting-type needles in 40 patients after spinal anaesthesia and found that the former needles were associated with a greater reduction in mean hearing level. A noncutting needle is thought to cause less damage to the dural fibres than a cutting needle and hence might reduce the leak of CSF after the procedure .
Age may be one of the risk factors of the occurrence of hearing loss. Gültekin and Ozcan  compared the incidence of hearing loss after spinal anaesthesia in young and elderly patients and found that the risk was much greater for the former. But two other studies of young patients failed to show the same significance [17,24]. Wang et al. studied 14 male patients (median age 65 years) undergoing transurethral resection of the prostate under spinal anaesthesia. They reported significant changes in hearing of greater than 10 dB in six out of 14 patients in the low-frequency ranges . The patients' mean ages were 38.1, 36.9 and 36.7, respectively, in the present study.
Several authors have suggested an association between PDPH and hearing loss [2,6,8,18], whereas others described hearing loss without associated headache [7,9]. The noncutting needles are considered to be the gold standard in spinal anaesthesia, not only because of the lack of cutting dural fibres but also because of a smaller rate of manufacturing flaws and less tip damage after bony contact, both of which result in an infrequent incidence of PDPH compared with sharp-bevel needles . The ballpen needle (noncutting needle) was designed to provide a combination of good puncture conditions with early identification of CSF along with an infrequent incidence of PDPH . In our study, when group B and group P were compared for change in hearing, no statistically significant difference was detected at any frequency tested. Two patients from group Q complained of PDPH. PDPH was not reported with the noncutting needles (groups B and P).
In conclusion, it seems that the type of spinal needle is one of the risk factors for hearing loss in patients who undergo spinal anaesthesia. Our study suggests that the use of ballpen and pencil-point needles reduces hearing loss after spinal anaesthesia compared with the cutting-type needle. Noncutting needles should be preferred.
1 Panning B, Mehler D, Lehnhardt E. Transient low-frequency hypoacousia after spinal anaesthesia
. Lancet 1983; 2:582.
2 Lee CM, Peachman FA. Unilateral hearing loss
after spinal anaesthesia
treated with epidural blood patch. Anesth Analg 1986; 65:312.
3 Walsted A, Salomon G, Olsen KS. Low frequency hearing loss
after spinal anaesthesia
: perilymphatic hypotonia? Scand Audiol 1991; 20:211–215.
4 Kilickan L, Gürkan Y, Ozkarakas H. Permanent sensorineural hearing loss
following spinal anaesthesia
. Acta Anaesthesiol Scand 2002; 46:1155–1157.
5 Cosar A, Yetiser S, Sizlan A, et al
. Hearing impairment associated with spinal anaesthesia
. Acta Otolaryngol 2004; 124:1159–1164.
6 Day CJE, Shutt LE. Auditory, ocular and facial complications of central neural block: a review of possible mechanisms. Reg Anesth 1996; 21:197–201.
7 Fog J, Wang LP, Sundberg A, Mucchiano C. Hearing loss
after spinal anaesthesia
is related to needle size. Anesth Analg 1990; 70:517–522.
8 Wang LP, Fog J, Bove M. Transient hearing loss
following spinal anaesthesia
. Anaesthesia 1987; 42:1258–1263.
9 Sundberg A, Wang LP, Fog J. Influence of hearing of 22 G Whitacre and 22 G Quincke needles. Anaesthesia 1992; 47:981–983.
10 Standl T, Stanek A, Burmeister MA, et al
. Spinal anaesthesia
performance conditions and side effects are comparable between the newly designed Ballpen and the Sprotte needle: results of a prospective comparative randomized multicenter study. Anesth Analg 2004; 98:512–517.
11 Panning B, Lehnhardt E, Mehler D. Transient low-frequency hearing loss
following spinal anaesthesia
. Anaesthesist 1984; 33:593–595.
12 Karatas E, Göksu S, Durucu C, et al
. Evaluation of hearing loss
after spinal anaesthesia
with otoacoustic emissions. Eur Arch Otorhinolaryngol 2006; 263:705–710.
13 Kiliçkan L, Gürkan Y, Aydin O, Etiler N. The effect of combined spinal-epidural (CSE) anaesthesia and size of spinal needle on postoperative hearing loss
after elective caesarean section. Clin Otolaryngol Allied Sci 2003; 28:267–272.
14 Wang LP. Sudden bilateral hearing loss
after spinal anaesthesia
: a case report. Acta Anaesthesiol Scand 1986; 30:412–413.
15 Walsted A. Effects of cerebrospinal fluid loss on the auditory system. Dan Med Bull 1998; 45:268–281.
16 Schaffartzik W, Hirsch J, Frickmann F, et al
. Hearing loss
after spinal and general anaesthesia: a comparative study. Anesth Analg 2000; 91:1466–1472.
17 Finegold H, Mandell G, Vallejo M, Ramanathan S. Does spinal anaesthesia
cause hearing loss
in the obstetric population? Anesth Analg 2002; 95:198–203.
18 Michel O, Brusis T. Hearing loss
as sequel of lumbar puncture. Ann Otol Rhinol Laryngol 1992; 101:390–394.
19 Michel O, Brusis T, Loennecken I, Matthias R. Inner ear hearing loss
following cerebrospinal fluid puncture: a too little appreciated complication? HNO 1990; 38:71–76.
20 Gültekin S, Ozcan S. Does hearing loss
after spinal anaesthesia
differ between young and elderly patients? Anesth Analg 2002; 94:1318–1320.
21 Wang LP, Magnusson M, Lundberg J, Tornebrandt K. Auditory function after spinal anaesthesia
. Reg Anesth 1993; 18:162–165.
22 Ready LB, Cuplin S, Haschke RH, Nessly M. Spinal needle determinants of rate of transdural fluid leak. Anesth Analg 1989; 69:457–460.
23 Malhotra SK, Joshi M, Grover S, et al
. Auditory function following spinal analgesia: comparison of two spinal needles. Eur J Anaesthesiol 2002; 19:69–72.
24 Ok G, Tok D, Erbuyun K, et al
. Hearing loss
does not occur in young patients undergoing spinal anaesthesia
. Reg Anesth Pain Med 2004; 29:430–433.
25 Parker RK, White PF. A microscopic analysis of cut-bevel versus pencil-point spinal needles. Anesth Analg 1997; 85:1101–1104.