Confirmation of Loss-of-Resistance for Epidural Analgesia De Q. H. Tran, MD, FRCPC,* Andrea P. González, MD,† Francisca Bernucci, MD,† and Roderick J. Finlayson, MD, FRPC* (Reg Anesth Pain Med 2015;40: 166–173) S uccessful epidural analgesia requires successful identifi- cation of the epidural space. Methods to recognize the lat- ter fall broadly into 3 categories: loss of resistance (LOR; tactile end point), negative-pressure recognition (visual end point), and acoustic fall in tonal pitch (auditory end point). Described in 1921 by Sicard and Forestier, 1 LOR remains the method most com- monly used. 2,3 In contrast, despite the many devices aimed at de- tecting 4–9 or augmenting negative pressure, 10–12 fewer operators rely on the latter to identify the epidural space. 3 For instance, the prevalence of use of the “hanging drop” can be as low as 11.7%. 13 Although works by Lechner et al 14–16 suggest that the fall in tonal pitch associated with needle transition from ligamentum flavum to epidural space can assist with catheter inser- tion, the sophisticated equipment required (pressure transducer, pressure amplifier, voltage-controlled oscillator, loudspeaker) limits its routine implementation. THE SHORTCOMINGS OF LOSS OF RESISTANCE Despite its popularity, LOR is far from perfect. Multiple articles have tackled the medium (air or saline). 17 Similarly, multiple improvements have been proposed to detect LOR (eg, pressurized syringe, microdrip method, continuous hydrostatic pressure system). 18–20 However, a potential drawback afflicting LOR may be its lack of specificity. For instance, age-related cysts can develop in interspinous ligaments and yield a false-positive LOR (LOR outside the epidural space). 21 At necropsy, up to 85% of subjects (aged 61–79 years) displayed an interspinous cyst in the lumbar region. 22 To complicate matters further, if the needle were to deviate from the midline, a nonepidural LOR could also occur upon entering paravertebral muscles. 23 In addition, cadav- eric specimens have demonstrated frequent midline gaps in ligamentum flavum in the cervical, upper thoracic (T1-T3), lower thoracic (T9-T12), and lumbar spine. 24–26 The presence of such gaps hinders the use of (midline) LOR because the latter no longer represents the interface between ligamentum flavum and epidural space; instead, it simply detects the crossing point between interspinous ligament and ligamentum flavum. Finally, thoracic paravertebral spaces and intermuscular planes (eg, be- tween the quadratus lumborum and psoas muscles) can also generate false-positive LORs. 27,28 These 2 confounders be- come problematic if the initial puncture site is excessively lateral and the transverse process mistaken for the lamina dur- ing a paramedian approach. PRIMARY FAILURE OF EPIDURAL ANALGESIA AND INCIDENCE OF FALSE-POSITIVE LOR Reasons for inadequate epidural blocks include primary fail- ure (incorrect placement of the epidural catheter) and secondary failure (catheter dislodgement/migration, suboptimal dosing of local anesthetic [LA] agents). 29 In turn, primary failure can be ascribed to misidentification of the epidural space or suboptimal catheter position. Primary epidural failure rates for labor and surgical analgesia vary between 2.0% and 16.9% in large stud- ies (n > 200) published over the last 10 years (2005–2014) 30–33 (Table 1). Because prudence dictates that LA boluses be frag- mented and administered through the catheter (and not the nee- dle), it can be difficult to ascertain if these reported failure rates stem from misidentification of the epidural space (nonepidural LOR) or catheter malposition (eg, intervertebral foramen). How- ever, because a variety of catheter tip locations are compatible with adequate LA response, 34 false-positive LOR may constitute a significant occurrence. 35 In summary, the contemporary liter- ature pertaining to labor and surgical analgesia does not permit definitive conclusions; in the worst-case scenario, false- positive LORs may occur in as many as 17% of patients. Two studies in chronic pain management can shed additional insight on nonepidural LORs. In 1985, Mehta and Salmon 36 car- ried out a series of 100 epidural blocks (85% lumbar, 9% cervi- cal, 6% thoracic). After LOR, the authors injected 0.1 to 0.2 mL of contrast (iophendylate) through the needle prior to obtaining radiographic confirmation. Mehta and Salmon 36 observed that, in 17% of cases, the needle tip (visualized by the spread of con- trast) was positioned either partially or completely outside the epidural canal. In 2001, Liu et al 37 performed 100 lumbar epi- dural blocks for patients with low-back pain. After obtaining LOR, the authors injected 4 mL of nonionic radiologic contrast through the needles: they observed an 8.3% incidence of false- positive LOR. 37 Although it might be tempting to infer from Mehta 36 and Liu’s 37 findings that the rate of nonepidural LORs would also be 8.3% to 17% in clinical anesthesia, such a parallel should be entertained with caution as spinal anatomy differs be- tween surgical/obstetrical patients and subjects with chronic spinal pain. Furthermore, Liu et al 37 used a smaller epidural needle (20-gauge), which may have affected tactile feedback for tissue planes. ADJUNCTS TO LOR The documented existence of nonepidural LORs argues for adjunctive modalities. In the literature, multiple adjuncts have been proposed. For the purpose of this article, only established modalities (radiographic imaging, test dose, paramedian ap- proach, dural puncture) or those with confirmatory data (wave- form analysis, electrical stimulation [ES], ultrasonography [US]) are discussed. Confirmatory data are defined as the pres- ence of more than 1 study investigating the adjunct and pub- lished in the English language. However, for the sake of completeness, anecdotal adjuncts are listed in the Appendix. From the *Department of Anesthesia, Montreal General Hospital, McGill University, Montreal, Quebec, Canada; and †Department of Anesthesia, Hospital de Carabineros, Santiago, Chile. Accepted for publication December 11, 2014. Address correspondence to: De Q. H. Tran, MD, FRCPC, Department of Anesthesia, Montreal General Hospital, 1650 Ave Cedar, D10-144, Montreal, Quebec, Canada H3G 1A4 (e‐mail: de_tran@hotmail.com). The authors declare no conflict of interest. Copyright © 2015 by American Society of Regional Anesthesia and Pain Medicine ISSN: 1098-7339 DOI: 10.1097/AAP.0000000000000217 DARING DISCOURSE 166 Regional Anesthesia and Pain Medicine • Volume 40, Number 2, March-April 2015 Copyright © 2015 American Society of Regional Anesthesia and Pain Medicine. Unauthorized reproduction of this article is prohibited.