The recent increase in recognition of nonarthritic hip conditions, such as labral tears and femoroacetabular impingement, has led to greater awareness of the potential overlap in symptoms from the spine and hip.3 Thus, in cases of ambiguous lower extremity pain, the potential involvement of the hip must be considered in the modern spine surgeon's clinical decision making.
Radiologic findings are often minimal and are difficult to assign diagnostic value to; physical examination findings can produce a confused and inconsistent view of the patient's condition.4 To avoid delay in the proper diagnosis and treatment of patients with complex hip-spine syndrome, additional testing may be required. Intra-articular injection of local anesthetic with or without steroid has been shown to be useful both in short-term pain relief and in determining the contribution of a specific joint to a patient's symptoms.5–10 It is a simple and undemanding test with high reported values for sensitivity and specificity in discerning the source of pain for patients with atypical hip pain or concomitant hip and spine disorders.11–13 This tool has been well studied in other joints, such as the shoulder,14 but a unified perspective of the literature on diagnostic hip injection to differentiate between hip and spine pathology has not been described.
The question being asked for this review is “Among patients with hip-spine syndrome specific to osteoarthritis of the hip and lumbar spine disorders, is intraarticular injection (ultrasound-guided or fluoroscopy-guided) a reliable diagnostic technique for determining pain source?” The purposes of this systematic review were to evaluate the literature regarding guided hip intra-articular injection used to identify the source of pain in the setting of both hip arthritis and lumbar spine disorders and to determine the diagnostic power and reliability of this tool.
Materials and Methods
A systematic review of the available literature was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.15
Studies meeting the following inclusion criteria were eligible for the review: sample size of more than 10 patients; adult patients older than 18 years; patients with hip-spine syndrome with concomitant hip and lumbar spine disorders; radiographs, ultrasound, computed tomography, or fluoroscopically guided intra-articular hip injections; clinical outcomes data; human studies; levels I, II, III, and IV evidence (according to the Oxford Centre for Evidence Based Medicine)16; articles published only in English; and publication in print or electronic journals. The exclusion criteria were as follows: review articles, basic science studies (except review article references), commentaries, case reports, cadaveric studies, or nonhuman studies; studies concerning epidural, intramuscular, or other injections besides hip intraarticular; studies regarding unguided intra-articular hip injections or injections guided only by anatomical landmarks; and studies with patients with no concomitant hip arthritis and lumbar spine disorders.
Two reviewers (D.R.M., B.H.M.) independently conducted database searches of both PubMed and the Cochrane Library in August 2018. The full texts of potentially eligible studies were obtained and reviewed for inclusion. The bibliography of each study was also reviewed to identify any additional literature. Differences between the 2 reviewers were discussed, and a consensus was reached.
The following search terms were used as MeSH and/or keywords: “osteoarthritis, hip” AND “injection, intra-articular”; “hip” AND “spine” AND “injection”; “osteoarthritis, spine” AND “injection, intra-articular”; “lumbar, vertebrae” AND “injection, intra-articular”; “low back pain” AND “injection, intra-articular”; “hip-spine disease” AND “injection, intraarticular”; “hip-spine syndrome” AND “injection, intra-articular”; and “lumbar spine disease” AND “injection, intra-articular.”
Database searches were performed by 2 independent reviewers (D.R.M., B.H.M.) using the inclusion and exclusion criteria. After removal of duplicates, the inclusion and exclusion criteria were applied to identify eligible articles. Two researchers (D.R.M., B.H.M.) independently screened titles and abstracts of the potentially relevant articles. Studies that met all criteria after this screening were then examined for full-text review. After full-text review, the final articles were selected for this systematic review.
Data were collected and recorded in a piloted computer spreadsheet program (Excel 2013; Microsoft, Redmond, Washington). Extracted data included the following: authors, journal, country, year of publication, level of evidence, number of patients, study design, imaging modality used, injection product, sensitivity, specificity, positive predictive value, and negative predictive value.
Two reviewers (D.R.M., B.H.M.) independently assessed the methodological quality of all included studies according to the Methodological Index for Non-Randomized Studies (MINORS) criteria.17 All information used in this review was extracted from the articles. Risk of bias was assessed for each study. Levels of evidence were collected from each study.
Synthesis of Results
Means for sensitivity (SN), specificity (SP), positive predictive value (PPV), and negative predictive value (NPV) among the studies, weighted by sample size, were calculated.
From the PubMed and Cochrane database searches, 397 and 38 studies were identified, respectively (Figure 1). Thirty-five studies were eligible for full-text review and 311 studies were excluded: 1 cadaveric study, 4 case report studies, 2 technical note studies, 11 review studies, and 293 irrelevant studies. After full-text review, 26 studies were excluded for the following reasons: 11 studies performed guided intra-articular injections in patients with hip arthritis but no lumbar spine disorders (patients with lumbar spine problems were excluded prior to hip injection) and 15 studies performed guided injections in patients with lumbar spine disorders but without hip arthritis. The study selection process is summarized in Figure 1.
Flow diagram showing the selection of studies for inclusion.
The systematic review search retrieved 9 studies that were conducted between 1991 and 2016: 4 prospective studies and 5 retrospective studies, with 4 of the studies conducted in the United Kingdom and 5 in the United States. Table 1 organizes each study's reported authors, journal, country study location, year, level of evidence, number of patients, study design, injection product, SN, SP, and PPV and NPV.
Regarding ethical considerations, 1 study recorded obtained written consent from each patient,18 3 studies received ethical approval from an institutional review board prior to being conducted,11,12,19 and the remaining studies did not cite ethical considerations. None of the studies reported potential conflicts of interest. A total of 697 patients were involved. The studies were divided into 2 groups according to the diagnostic injection product used for intra-articular hip injection: studies that used only anesthetic (bupivacaine, lidocaine, bupivacaine hydrochloride, or ropivacaine) were placed in one group (anesthetic only group) and studies that used anesthetic with corticosteroids were placed in a second group (anesthetic with corticosteroids group). The anesthetic only group included 6 studies with a total of 363 patients6,10,12,18–20 and the anesthetic with corticosteroids group included 3 studies with a total of 334 patients.8,11,13
All studies included in this systematic review were nonrandomized. The MINORS instrument21 was used to assess the methodological quality of each study. Mean score for the studies included in this systematic review was 12.33 of 16 points. Scores for nonrandomized and noncomparative studies can be understood as very low quality (0 to 4); low quality (5 to 8); moderate quality (9 to 12); and high quality (13 to 16).
Results of Individual Studies
The anesthetic only group included 6 studies and 363 patients.6,10,12,18–20 Two studies12,19 were retrospective, and 4 studies6,10,19,20 were prospective. One of the common objectives was to use guided intra-articular hip injection to differentiate pain source in patients with both hip and lumbar spine arthritis. Physical examination and proper hip and lumbar spine radiography protocols were used to diagnose simultaneous findings of arthritis in all studies.
Five studies with a combined 321 patients reported patients' sex, of which 65.73% were female.6,12,18–20 The same 5 studies reported patient age, and the mean age weighted by the studies' sample sizes was 51.74 years.6,12,18–20 Studies described 5 different protocols for administration of the anesthetic injection.
One study used 10 mL of 0.25% bupivacaine,18 1 used 10 mL of 0.5% bupivacaine,20 1 used 3.5 mL of 2% lidocaine plus 3.5 mL of 0.5% bupivacaine,12 2 used 10 mL of 0.5% bupivacaine,6,10 and 1 used a median of 5 mL of 0.5% ropivacaine, 0.25% bupivacaine, or 1.0% lidocaine, depending on anatomical constraints.19 For intra-articular localization, 6 studies used fluoroscopy-guided injections,6,8,11–13,20 1 used radiographs,10 1 used ultrasound,18 and 1 used fluoroscopy or ultrasound.19
Positive response after guided intraarticular hip injection was specifically defined in 3 studies. One study defined positive response as 70% or greater relief of pain11; 1 study took 50% pain relief from the pre-injection pain after 30 minutes or less12; and 1 study grouped patients into achieving more or less than 50% pain relief 24 hours after the injection.19 The remaining studies did not use any specific number or parameter to define positive response.
In the anesthetic with corticosteroids group, 3 retrospective studies were included with a combined total of 334 patients.8,11,13 Two hundred (59.9%) patients were women and 134 (40.1%) were men. Mean age was provided by all of the studies, and the overall mean weighted by sample size was 63.6 years. The 3 studies aimed to determine the source of hip pain in patients with both hip and lumbar spine arthritis (hip-spine syndrome).8,11,13 Three studies used physical examination and proper hip and lumbar spine radiography protocols to diagnose patients with simultaneous findings of hip and lumbar spine arthritis.8,11,13 Different protocols were used in each study for intra-articular diagnostic hip injection. One study used 5 mL of 0.5% bupivacaine plus 1 mL (80 mg) of methylprednisolone,11 1 study used 9 mL of 1% lidocaine plus 1 mL (40 mg) of triamcinolone,13 and 1 study used 0.5% of bupivacaine plus an unspecified quantity of triamcinolone.8
Regarding guided technique, fluoroscopy was used in all of the studies. Only 1 study mentioned specifically how positive response after intra-articular hip injection was assessed,11 considering 50% pain relief from the pre-injection pain after 30 minutes a “positive response” and 50% pain improvement within 2 weeks following the injection a “delayed positive” that was also included in the “positive group” for results. Values of SN, SP, PPV, and NPV were provided and extracted from each study.
Synthesis of Results
In the anesthetic only group, 3 studies reported SN, SP, PPV, and NPV for guided intra-articular hip injection. One study reported 97%, 90.9%, 97%, and 90.9% for SN, SP, PPV, and NPV, respectively20; 1 study reported 95%, 87.50%, 95%, and 87.50% for SN, SP, PPV, and NPV, respectively12; and 1 study reported 88%, 100%, 100%, and 33% for SN, SP, PPV, and NPV, respectively.10
Another study reported PPV of 95.2% and NPV of 87.5%.12 One study only reported SN of 96%.6 The sixth study did not report SN, SP, PPV, or NPV.18 Weighted by sample size, the mean values for the anesthetic only group were 95.1%, 93.4%, 97.2%, and 78.1% for SN, SP, PPV, and NPV, respectively. No major complications were reported regarding infiltration itself. One study reported 1 minor vasovagal reaction during the procedure, but the patient completely recovered by the conclusion of the procedure.12 In the anesthetic with corticosteroids group, 1 study reported values of 91.5%, 100%, 100%, and 84.6% for SN, SP, PPV, and NPV, respectively.11 The second study reported values of 100%, 81%, 97%, and 100% for SN, SP, PPV, and NPV, respectively.13 The third study showed values of 88%, 100%, 100%, and 85% for SN, SP, PPV, and NPV, respectively.8 When the studies were combined, the mean values were 93.1%, 95.3%, 99.3%, and 88.5% for SN, SP, PPV, and NPV, respectively. No infections or neurovascular complications were reported.
Among all studies in this review, the mean reported values weighted by sample size were 93.6%, 95.0%, 98.8%, and 86.3% for SN, SP, PPV, and NPV, respectively. The values were comparable between groups.
This systematic review identified 9 level IV studies reporting the efficacy of guided intra-articular injection of the hip in distinguishing between hip and spine-derived pain. Three studies used corticosteroid in addition to anesthetic,8,11,13 and 6 used anesthetic alone.6,10,12,18,19,20 Six studies used fluoroscopic guidance,6,8,11–13,20 1 used ultrasound,18 1 used either fluoroscopy or ultrasound,19 and 1 used radiographs.10 High values were reported for SN, SP, PPV, and NPV in identifying the hip as the source of lower extremity pain. Weighted by sample size, the mean values for SN, SP, PPV, and NPV were 93.6%, 95.0%, 98.8%, and 86.3%, respectively.
Hip pathology often occurs in combination with lumbar stenosis and back pain, which can make diagnosis and treatment for patients with these disorders difficult because symptoms of hip and spine disorders often overlap.22 First introduced by Offierski and MacNab,1 the term hip-spine syndrome is used to describe patients who have concomitant hip and lumbar spine disorders. In simple cases of hip-spine syndrome, a standard assessment that includes a careful history, physical examination, and plain radiographs of the hip and spine is sufficient to identify whether the patient's symptoms are caused primarily by hip or spine pathology.7–9
However, in complex cases of hip-spine syndrome, it is difficult to determine the source of pain, and additional diagnostic testing must be performed.3,4,23,24 Guided diagnostic intra-articular hip injections serve as a reliable tool that can be used to clarify a patient's diagnosis and provide guidance for appropriate therapeutic options.25–30 When diagnostic injections are performed without image guidance, they are often unreliable; up to 40% of blind hip injections are administered into the extra-articular space.31,32 The aim of this systematic review was to evaluate the role and importance of guided diagnostic intra-articular hip injection in determining the pain source in the setting of complex hip-spine syndrome.
Previous studies have examined the use of intra-articular injection to identify various hip joint pathologies. A systematic review by Khan et al3 reported that intraarticular hip injection is a useful tool in the diagnosis, therapy, and prognostication of femoroacetabular impingement. In the setting of other intra-articular hip pathologies such as labral tears and ligamentum teres ruptures, fluoroscopically guided intra-articular hip joint injections demonstrated a 90% accuracy rate in a study by Byrd and Jones.25
Diagnostic injections have also been used in other joints to differentiate between concomitant pathologies. Historically, this diagnostic tool has been used to differentiate between shoulder and cervical spine pathologies.14 The use of guided intraarticular hip diagnostic injection was first reported by Kleiner et al10 for the purpose of distinguishing between hip and referred lumbar spine pain in patients with coexisting hip and spinal arthritis. They reported a SN of 88% and a SP of 100%, comparable to the data provided by other studies included in the current review. The use of epidural injections for lumbar pathology in the setting of hip-spine syndrome is less well defined in the current literature.19,26 Although the use of anesthetic and steroid combination has not been shown to be significantly more accurate, it has been reported to relieve pain for longer (up to 1 year) compared with an injection with only anesthetic (up to 1 week).21,32–35
Ultrasound-guided injections offer the following advantages compared with fluoroscopy-guided injections: no ionizing radiation exposure, greater accessibility, smaller equipment, greater accuracy of delivery, and the possibility of visualizing soft tissue structures.25,30 Yoong et al18 stated that the use of ultrasound is potentially more cost-effective and faster than the use of fluoroscopy and that it avoids inadvertent puncture of femoral vessels. Furthermore, Byrd and Jones25 reported that in-office ultrasound-guided injections of the hip were more convenient and less painful than fluoroscopic-guided, hospital-based injections.
This study followed the validated PRISMA method to conduct an exhaustive search of available literature relevant to the topic. Using multiple databases helped to ensure that the best-available evidence was presented. Combining the efforts of 2 independent reviewers limited the potential effects of errors or ambiguity in selecting articles for inclusion. Finally, evaluating according to the MINORS criteria confirmed the scientific rigor of the reviewed studies. The average MINORS score for the included studies was 12.33, indicating a high level of quality of evidence.
The authors acknowledge several limitations to this study. From a methodological standpoint, this review lacked evidence from high-quality randomized controlled trials. The studies presented in this review were all level IV and lacked a control group. In addition, the number of studies included was relatively low and many of these studies reported on relatively small samples. Another limitation stems from the fact that there were important variations between the intraarticular injection protocols used, the definition of a positive response to injection, and the tool used for guidance (ultrasound vs fluoroscopy).