(0148-396X/84/1406-0697802.00/0, Nevkosunciay Vol, 14, No.6, 1984 Paned in USA Copyright© 1984 by the Congress of Neurological Surgeons Dexamethasone—a Helpful Adjunct in Management after Lumbar Discectomy J. Stovall King, M.D. Carolina Neurosurgical Service, Florence, South Carolina ‘This randomized prospective study was designated to test the hypothesis that intraoperative and postoperative administration of a potent antiinflammatory steroid (dexamethasone) would reduce patients’ postoperative pain. The hypothesis is based on the concept that inflammation is associated with a lowering of the threshold for nociceptive sense ‘organs and thatthe inflammatory process is responsible, atleast in part, for postoperative pain. The quantity of narcotics requested by the patients during the first 72 hours of the postoperative period was used as the measure of their ‘postoperative pain, Patients treated with steroid who underwent lumbar laminotomy and discectomy used considerably less narcotic during the postoperative period than those not treated with steroid (P value < 0.01). Less difference was noted in patients who underwent laminectomy, and no statistically significant difference was noted for those who underwent anterior discectomy. It is concluded that the use of an antiinflammatory steroid during and after operation, significantly reduces the immediate postoperative pain after lumbar discectomy and may be useful in the postoperative ‘management of other surgical procedures. (Neurosurgery 14:697-700, 1984) Key words: Antinflammatory steroid, Dexamethasone, Laminectomy, Lumbar laminotomy and discectomy, Narcotic medication, Nociceptive sense organ, Postoperative pain INTRODUCTION During a series of animal experiments in which recordings were made from single cutaneous C fibers, it was concluded that the threshold oft least one type of cutaneous polymodal nociceptor is lowered by certain endogenous chemicals liber- ated during the inflammatory process (5). Among the candi- date chemicals involved are histamine, bradykinin, seratonin, and prostaglandin. As the threshold is iowered, nociceptor are excited by stimuli that previously had been subthreshold for these receptors. It follows that a reduction ofthe inflammatory ‘process might prevent or restrain the alterations in nociceptor threshold that take place when tissue is damaged. This study was designed to test the hypothesis that intraoperative and postoperative administration of a potent antiinflammatory Steroid (dexamethasone) could interfere with a process that results in the lowering of nociceptor thresholds during surgical 0.2; P value using the Wileoxon test, >0.2),698 KING ‘The results for patients who underwent lumbar lami- notomy and discectomy are presented in Figure 3, and the ‘characteristics ofthe control and treated groups are compared in Table 2, For this operation. 13 patients were in the control group and 15 were in the treated group. Patients who did not 109, 3 a0] f BE ° | Cerven g8 Asses Fg 00 | 28 | 83 ge ge) 5 si 368 | 23 20 3 ° ‘anterior Lumber Lumbar Cervical Laminectomy _Discectomy Discectomy Fic. 1. Average amount of postoperative analgesic requested by the control and steroid-teated groups Tor each type of operation studied, 8 Mean. ing of Methadone or Morphine Used During Post-operative Period O Nonstreated Treated Group ‘Group Fic. 2. Range and mean amounts of postoperative analgesic re- {quested by the control and steroid-reated patients who underwent anterior cervical discectomy, Neurosurgery, Vol. 14, No. 6 receive steroids used a much higher and statistically significant amount of morphine or methadone (difference, 48.3 mg: standard error of this difference, 16; P value, 0.01 for both the ‘test and the Wilcoxon test). Forty per cent of the steroid- treated patients who underwent lumbar discectomy required no narcotic analgesics during the first 72 hours of the post- ‘operative period, whereas none of the untreated group de- clined narcotics. These results were statistically conclusive ‘The most ambiguous results were obtained from those patients undergoing lumbar laminectomy. These results are presented in Figure 4, and the characteristics of the control 1d treated groups are compared in Table 3, These operations were all done for spinal stenosis, postoperative epidural scar- ring, or a lateral sulcus syndrome. For this operation, 13, patients were in the control group and 11 were in the treated ‘group. The non-steroid treated patients did use more analee- sics (although only marginally significant statistically) than the steroid-treated patients (Wilcoxon rank test P value, (0.0426; difference, 30.8 mg: standard error of this difference, 7.7; test P value, 0.054). This result i just barely statistically reliable based on a Wilcoxon rank test. The usual (testis just barely not significant. ‘The statistical evaluation of all results relied on two-sided tests, with the assumption that it was possible that the non- steroid group could have experienced less postoperative pain than the steroid group. If this hypothesis were rejected, then all P values would be divided by 2, and the results for the lumbar laminectomy subgroup would become conclusively significant statistical DISCUSSION It is well recognized that the use of antiinflammatory ste~ roids may significantly reduce the pain associated with lumbar 8 Wve Period 8 8 Mean & img of Methadone or Morphine Used During Post-ope Noxteated Treated ‘eeu Broup Fic. 3. Range and mean amounts of postoperative analgesic re- ‘quested by the control and steoid-treated patients who underwent lumbar laminoiomy and discectomy Tame 1 Patients Who Underwent Age (98) No, Cases Sex = Range Mean Conol—15. IM 34-05 46 4F Steroid: 15M 30-59 462 “treated —21 oF Anterior Cervical Discectomy. Operation Previous NO. Patents Who Received a Nedeaton Singe Mule “UST — Morphine Methadone Both Boe m0 4 om ttJune 1984 DEXAMETHASONE—A HELPFUL ADJUNCT — 699 ‘Tame 2 Patients Who Underwent Lumbar Discectomy No. Patients Who Received (yr) ‘ration Previous at ore Operation Medication No. Cas se Dreaion _Mteation | ange Mean —‘Sinb Mule Satine Methadone Both Control—13, 8M 24-58 427 13 0 3 10 o 3 Sr seri SMe ks 7 2 0 reacts TF : 7 Tatts 3 a Patents Who Undernent Lamar Lamineomy hee w peion Press NPs Wo Reh Noe so __ Geeston ato 10. Cases * at Same mame Man Sind Malibe 1ST ich Methadone Both GontelB a wi “oO or Steri SM 2670460 ee tenet ar ° so 100) In retrospect, the inclusion of anterior discectomy patients . in our study was inadvisable. So litle sue damage fine 23 valved inthis operation thatthe vast raj of paints Ee eo] leave the hospital on the Ist or 2nd postoperative day. So little analgesic was used during the postoperative period by ao teak groupe that's Suisealysgnifeant frees beeen 5 4°) vay the steroid-treated and the control groups would be difficult 2? to demonstrate a3 ‘The results for patients undergoing laminectomy were more ae” suggestive of a beneficial effect of postoperative steroids. af ‘Whether they are statistically concliive 1 debatable (see = Results), These vaults might have been more conclusive i e a larger group of patients had been studied or a higher dose FG. 4, Range and mean amounts of postoperative analgesic re ‘quested by the conteol and steroid-reated patients who underwent lumbar laminectomy. dise rupture (1-3, 6, 8-10). Some authors have argued that treatment with steroid is so effective that it should be used instead of surgical intervention for lumbar disc herniation (4) Tie only report of which lam aware regarding postoperative steroid use in treating patients who have undergone lumbar discectomy is by Naylor etal. (7). They examined the “pos- sible beneficial effets of dexamethasone on the resolution of signs and symptoms in postdisc surgery.” The focus of their attention was on the first 3 postoperative months. They evaluated the amount of residual postoperative pain, the residual restriction of spinal flexion, the degree of straight leg raising the return of reflexes, residual sensory changes, and ‘motor weakness. Looking at the problem from this perspec- tive, they concluded “that dexamethasone has no significant effect on the postoperative course in patients subjected to laminectomy for dise prolapse.” Buried in the report, how- ever isthe statement that “the dexamethasone group required less analgesic drug therapy in the first seven days.” This, observation was corroborated by the present study of dexamethasone had been used for the treated patients. The statistical data for the laminotomy-discectomy group, are conclusive and certainly agree with our clinical observa tions during the trial. Forty per cent of the steroid-treated patients who underwent lumbar discectomy required no nar- cotic analgesics during the first 72 hours of the postoperative period, whereas none of the untreated group declined narcot- ies. It was my clinical impression that more of the steroi ‘weated group than the control group were walking in the halls ‘on their Ist postoperative day. However, this observation was not quantitatively measured, The euphoria that so often ac- companies the use of steroids also added to the gratifying immediate postoperative course of patients so treated while in the hospital. No unusual temperature elevations, hyperten- sion, or gastrointestinal distress was noted in the steroi treated group. No patients experienced a clinically significant delay in wound healing, and none of the patients in either group developed a postoperative infection, CONCLUSION ‘The use of dexamethasone during and in the first 72 hours. after operation significantly reduces the immediate postoper- ative pain following lumbar laminotomy and discectomy and may reduce the postoperative pain following other surgical procedures.700 KING ACKNOWLEDGMENTS 1 thank R. J. Carroll, Ph.D., Professor, Department of Statistics, University of North Cavolina for statistical evalua- tion of the data. I also thank Irene Coley, P.A., for her assistance in collecting the data and Ray Kobler for her preparation of the manuscript. Received for publication, June 1984) Reprint requests: J. Stovall King, M.D, Carolina Neurosurgical Service, $06 East Cheves Stree, Suite 206, Florence, South Carolina 29501 |. 1983; accepted, February 10, REFERENCES 1. Barry PIC, Hume Kendall P: Corticosteroid infiltration of the extradural space, Ann Phys Med 6:267-273, 1962, 2. Dilke TFW. Bury HC, Grahame R: Extradural corticosteroid nection in management of lumbar nerve root compression. Br Med J 2:635-637, 1973. 3, Feller HI: Treatment of low-back and scatie pain by injection of hydrocortisone into degenerated intervertebral discs. J Bone Joint Surg [Am] 38A:585-592, 1956. 4. Green LN: Dexamethasone in the management of symptoms due to herniated lumbar dise. J Neurol Neurosurg Psychiatry 38:1211-1217, 1975, 5. King JS, Gallant P, Myerson V, Perl ER: The effets of antiin- flammatory agents on the responses and the sensitization of ‘unmyelinated (C) fiber polymodal nociceptors, in Zotterman Y. (ed): Sensory Functions of the Skin in Primates with Special Reference 10 Man (Wenner-Gren Center Intemational Sympo- sium Series, vol 28). Oxford, Pergamon Press, 1976, pp 441-461 6. Lindholm R, Salenius P: Caudal, epidural administration of anesthetics and corticoidsin the treatment oflow back pain, Acta Orthop Scand 35:144-165, 1968, Naylor A, Flowers MW, Bramley JED: The value of dexameth- asone in the postoperative treatment of lumbar disc prolapse, Orthop Clin North Am 8:3-8, 1977 8. Naylor A, Turner RL: ACTH in the treatment of lumbar disc prolapse. Proc R Soc Med $4:282-284, 1961. 9, Sehgal AD, Gardner WJ: Subarachnoid injection of cortcoste- roids in the management of sciatica. J Indian Med Assoc 39:291— 293, 1962. 10, Wilkinson HA, Schuman N: Intradiscal corticosteroids in the ‘treatment of lumbar and cervical dise problems. Spine $:385~ 389, 1980. COMMENTS, Dr. King has shown the beneficial effect of systemic corti costeroids as adjunctive pain control medications during the early postoperative period after lumbar spinal operation. Al- though I take some exception to basing.a statistical conclusion (on the milligrams of analgesics used when this lumps together milligrams of morphine and methadone, I doubt that this would materially alter the author's conclusions. Depository corticosteroids can also be helpful in similar patients when used intraoperatively. I can personally reaffirm observations made by others regarding the pain amelioration provided by the intraoperative use of depository corticoste- roids epidurally or at the bone donor site, Unfortunately, some of my patients experienced a rapid resurgence of pain Neurosurgery, Vol. 14, No. 6 10 to 14 days postoperatively, presumably as the depository corticosteroid becomes exhausted, ‘What the reader will need to weigh in Dr. King’s study is not only the relative efficacy, but the relative safety of corti costeroids vs, opiates during the postoperative period. I am strongly convinced that patients do better in the long run if they are made more comfortable with adequate medication early postoperatively to facilitate early return to activity and to avoid the fear of pain, but 1 wonder if corticosteroids in the short term use are more or less dangerous than opiates in short term use, Harold A. Wilkinson, M.D. Worcester, Massachusetts Many neurosurgeons have used a short course of cortico- steroid therapy in selected postdiscectomy patients to control carly postoperative pain. Dr. King is to be commended in attempting to put this practice on a sounder scientific base by demonstrating its clinical efficacy in a randomized, controlled tial, The positive conclusions of this study are useful to ‘neurosurgcons and should help others to reduce the pain and discomfort of patients during the early postoperative period, ‘The study is not large enough 0 allow conclusions to be drawn about the side effects ofthis treatment. The infection rate in these operations is so low that even a doubling of the infection rate in a steroid-ireated group, which would be considered clinically important, would be extremely difficult to detect without a very large number of patients, We also do not know whether the full 72 hours of treatment is necessary ‘or whether a much shorter course of steroid therapy would bbe equally effective. Many patients actually have less pain after laminectomy and discectomy than preoperatively and do not need to be given steroid. It would be useful to find some way of predicting which patients really require such treatment. Perhaps future studies will clarify these questions. ‘There are two minor criticisms of the technique of con- ducting the trial. The method of allocating patients to the treatment and control groupsis nota truly random procedure. Itis potentially open to abuse. There is no reason to question the validity of the allocation inthis study, but tables of random numbers and computer-generated pseudo-random numbers are so widely available that itis preferable to use such methods for truly random allocation. Second, in assessing a highly subjective outcome such as pain, current standards require double-blind evaluation. The potential for the unknowing introduction of bias into the treatment and evaluation of these patients is so great and so well documented that some blinding technique should be used whenever possible. In addition to the clinically useful information presented in this study, King has demonstrated that scientific clinical investigation need not be overly arduous and expensive and that it need not be conducted in large academic centers. He, and other neurosurgeons, are encouraged to continue to con- ‘duct such trials, which will help to put the clinical practice of neurosurgery on a sound scientific basis, Stephen J. Haines, M.D. Minneapolis, Minnesota