ORIGINAL ARTICLE

Evaluation of a Systematic Approach to Pediatric Back Pain: The Utility of Magnetic Resonance Imaging Norman Ramirez, MD, John M. Flynn, MD, Brian W. Hill, MD, Jose A. Serrano, MD, Carlos E. Calvo, MD, Rafael Bredy, MD, MBE, MScCR, and Raul E. Macchiavelli, PhD

Background: Several studies have suggested that back pain in the majority of pediatric patients does not have an identifiable cause. Many children undergo extensive diagnostic workup that ultimately results in a nonconfirmative diagnosis. The purpose of this study was to (1) describe the prevalence of back pain seen in a pediatric orthopaedic clinic; (2) evaluate the efficacy of a systematic approach dependent on magnetic resonance imaging (MRI) in the diagnosis of pediatric back pain; and (3) analyze sensitivity, specificity, positive predictive value, and negative predictive value of various clinical signs and symptoms. Methods: For a 24-month period, all patients that presented with a chief complaint of back pain were prospectively enrolled in this study and evaluated in a systematic approach which utilized MRI for patients with constant pain, night pain, radicular pain, or abnormal neurological examination after an initial history, physical examination, and negative radiographic examination. Results: The prevalence of chief complaint of back pain was 8.6% (261/3042 patients). Of the 261 patients, 34% had an identifiable pathology following the systematic approach. In 8.8% of patients, the diagnosis was established with the history, physical examination, and plain radiographs. MRI yielded a definitive diagnosis in another 25% of patients. It is noteworthy that of the 89 patients with a confirmed pathology, 26% were identified with plain radiographs and 74% with MRI. Conclusions: A systematic approach to diagnose pediatric back pain demonstrated that 34% of pediatric patients that present to an outpatient orthopaedic clinic complaining of back pain will have identifiable pathology. The diagnostic yield increased from 8.8% with the history, physical examination, and plain radiographs to 22% with the TCN Bone Scan to 36% with the use of the MRI. The clinician should be aware that the presences of lumbar pain or constant pain are red flags for the presence of underlying pathology. Level of Evidence: Level III. Key Words: pediatric back pain, algorithm, painful scoliosis (J Pediatr Orthop 2015;35:28–32)

From the Hospital de la Concepcion, San German, Puerto Rico. The authors declare no conflicts of interest. Reprints: Norman Ramirez, MD, Hospital de la Concepcion, San German, Box 6847 Mayaguez, 00681 Puerto Rico. E-mail: [email protected]. Copyright r 2014 by Lippincott Williams & Wilkins

28 | www.pedorthopaedics.com

T

he reported prevalence of pediatric back pain has been limited because this disorder was historically considered rare in children. Early studies of Turner et al1 and Grantham2 described a prevalence of back pain in the pediatric population between 2% and 11%. Recent research has reported a much higher prevalence between 27% and 51% with the use of a validated questionnaire. These studies showed a dramatic onset of symptomatic back pain that began at 10 years of age with a continued increase during teenage years.3–8 King9 and Hensinger10 reported that 66% to 84% of the patients have an evident etiology of back pain that require treatment. On the basis of their findings, the authors recommend an exhaustive investigative protocol in every pediatric patient with back pain to identify the possible etiology. These findings have undergone recent scrutiny and may reflect the type of practice or referral pattern. Despite the absence of neurological symptoms, children were often subjected to multiple studies, increased patient/family anxiety, and costly diagnostic workup that ultimately results in a nonconfirmative diagnosis.11–14 Several recent studies have been unable to find an identifiable cause in the majority of pediatric patients complaining of back pain.12–14 The need to find serious pathology is imperative, but the methodology has been controversial. Several authors have attempted to design a reasonable algorithm that can be followed to diagnose the cause of pediatric back pain using the most efficient diagnostic studies.12–14 It is widely accepted that a detailed history and physical examination in conjunction with plain posteroanterior (PA) and lateral standing roentgenogram of the entire spine are paramount for the initial diagnostic assessment of back pain. If the initial radiographs are positive, a specific diagnosis is documented and treatment can be initiated. There are diverging perspectives in the next step if the history, physical examination, and radiographs do not reveal a specific diagnosis. Some authors recommend a technetium bone scan,13–15 whereas others advocate a magnetic resonance imaging (MRI).16,17 In 2006, Feldman et al17 presented a systematic approach dependent on MRI in the diagnosis of pediatric back pain in those patients with negative history, physical examination, and radiographs. The purpose of this study was to: (1) describe the prevalence of back pain seen in a pediatric orthopaedic clinic; (2) evaluate the efficacy of a systematic approach dependent on MRI in the diagnosis of pediatric back pain J Pediatr Orthop



Volume 35, Number 1, January 2015

J Pediatr Orthop



Volume 35, Number 1, January 2015

Pediatric Back Pain Systemic Approach: MRI Utility

as proposed by Feldman et al17; and (3) analyze sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of various clinical signs and symptoms.

METHODS All pediatric patients (18 y of age and younger) who presented to Concepcio´n Hospital Orthopaedic clinic in San German, Puerto Rico between September 2006 and September 2008, were prospectively evaluated for the study. The inclusion criteria consisted of: (1) patients older than 4 years of age and younger than 18 years of age; (2) presenting with a chief complaint of back pain (> 4 wk duration); (3) without history of trauma; or (4) a previous definitive diagnosis and treatment. Patients were excluded if they had a definite known source of back pain, cervical pain, history of trauma, pain 10 y Constant Intermittent No night pain Night pain Normal neuro evaluation Abnormal neuro evaluation Duration < 3 mo Duration > 3 mo Lumbar pain Thoracic pain VAS < 5 VAS > 5 No scoliosis Scoliosis Cobb angle < 25 degrees Cobb angle > 25 degrees

P

Yes

No

Odds Ratio

50 39 8 81 83 6 78 11 64 25 31 58 79 10 29 53 63 26 82 7

127 45 28 144 57 115 136 36 171 1 39 133 146 26 60 94 108 64 159 13

0.454 (0.264-0.779) 1 0.508 (0.208-1.119) 1 27.909 (12.380-75.150) 1 1.877 (0.930-4.059) 1 1 66.67 (8.85-500) 1.833 (1.035-3.203 1 1.407 (0.663-3.199) 1 0.857 (0.488-1.490) 1 1.436 (0.833-2.519) 1 0.958 1

0.0042 0.0950 < 0.0001 0.0798 < 0.0001 0.0378 0.3820 0.5866 0.1945 0.9297

VAS indicates visual analog scale.

8.8% to 22% according to Feldman et al,17 Bhatia et al,12 and our study. If the initial history, physical examination, and radiographs do not reveal any specific diagnosis, the next step is disputable. Some studies recommend technetium bone scan,12–15 whereas other authors advocate MRI.7,16,17 With the use of bone scan as the next step, Feldman and colleagues and Bhatia and colleagues could increase the diagnostic yield by 4%, missing critical lesions such as spinal cord pathology. Subsequently, Feldman et al17 supplanted bone scan with MRI increasing the diagnostic yield 10%. In our study 8.8% received a diagnosis by history, physical examination, and radiographs. For an additional 25% of patients, an identifiable diagnosis was established by using MRI. In a larger patient population sample, we found 34% of patients evaluated to have identifiable pathology. This study confirms that pediatric back pain has a low rate of diagnosable pathology. Historically, the most common diagnoses were spondylolysis with or without spondylolisthesis, Scheuermann disease followed by infections, disk patho-

logy, and malignancies.1,2,22,23 Conversely, we found that disk pathology (17%) was the most common finding followed by spondylolysis/spondylolisthesis and Scheuermann disease. This change is most likely attributed to the use of MRI as the diagnostic tool, which is highly sensitive for disk pathology.12,16,17 The second most common pathology was spondylolysis with or without spondylolisthesis. If the x-ray is negative, there is a controversy about which is the best diagnostic test for spondylolysis.16 Several studies recognized that MRI could be used as the imaging technique, because the presence of bone edema is detected with great clarity on both sagittal STIR and fat-saturated T2 imaging.16 Other studies disagree and recommend the use of bone scan to confirm the diagnosis of spondylolysis.12,23,24 We prefer the use of MRI to avoid the high radiation exposure associated with the bone scan.25 We compare our findings with those of Feldman et al17 for the diagnostic tools (constant pain, night pain, abnormal neurological examination, lumbar location, and the presence of scoliosis) used in both the studies.

TABLE 2. Sensitivity, Specificity, Positive Predictive Value, and Negative Predictive Value of Various Clinical Signs and Symptoms to Establish a Specific Diagnostic Pathology Male Age < 10 y Constant pain Night pain Abnormal neuro evaluation Duration < 3 mo Lumbar VAS > 5 Scoliosis Cobb angle >25 degrees

Sensitivity (%)

Specificity (%)

Positive Predictive Value (%)

Negative Predictive Value (%)

44 9 93 12 28 35 89 65 29 8

74 84 67 79 99 77 15 39 63 92

46 22 59 23 96 44 35 36 29 35

72 64 95 64 73 70 72 67 63 66

VAS indicates visual analog scale.

r

2014 Lippincott Williams & Wilkins

www.pedorthopaedics.com |

31

J Pediatr Orthop

Ramirez et al

Constant pain and lumbar location were found to have higher sensitivities in both the papers. The presence of night pain, abnormal neurological examination, and scoliosis are clinical findings that are not likely to detect underlying pathology and were found to be less sensitive in both studies. In summary, when we compare our results with those described by Feldman et al, there are only minor statistical differences between both studies regarding the aforementioned signs and symptoms. In conclusion, this study establishes that 8.6% of pediatric patients who present to a pediatric orthopaedic clinic will complain of back pain and 34% of them will have an identifiable pathology. Our results support the use of a systematic algorithm that utilizes MRI as an efficient, well-organized methodology to evaluate back pain in the pediatric population. The clinician should be aware that the presences of constant pain or lumbar pain are red flags for the presence of underlying pathology. REFERENCES 1. Turner PG, Green JH, Galasko CS. Back pain in childhood. Spine. 1989;14:812–814. 2. Grantham VA. Backache in boys—a new problem? Practitioner. 1977;218:226–229. 3. Olsen T, Anderson R, Dearwater S, et al. The epidemiology of low back pain in an adolescent population. Am J Public Health. 1992;82:606–608. 4. Balague F, Skovron M, Nordin M, et al. Low back pain in schoolchildren. A study of familial and psychological factors. Spine. 1995;20:1265–1270. 5. Balague F, Dudier J, Nordin M. Low-back pain in children. Lancet. 2003;361:1403–1404. 6. Fairbank J, Pynsent P, Van Poortvliet J, et al. Influence of anthropometric factors and joint laxity in the incidence of adolescent back pain. Spine. 1984;9:461–464. 7. Bockowski L, Sobaniec W, Kulak W, et al. Low back pain in school-age children: risk factors, clinical features and diagnostic management. Adv Med Sci. 2007;52(suppl 1):221–223. 8. Troussier B, Davoine P, de Gaudermaris R, et al. Back pain in school children. A study among 1178 pupils. Scand J Rehabil Med. 1984;26:143–146.

32 | www.pedorthopaedics.com



Volume 35, Number 1, January 2015

9. King HA. Evaluating the child with back pain. Pediatr Clin North Am. 1984;31:1083–1095. 10. Hensinger RN. Back pain in children. In: Bradford DS, Hesinger RN, eds. The Pediatric Spine. New York: Thieme; 1985:41–60. 11. David J. Evaluation of back pain in children. Pediatr Child Health. 2007;18:56–60. 12. Bhatia N, Chow G, Timon S, et al. Diagnostic modalities for the evaluation of pediatric back pain. A prospective study. J Pediatr Orthop. 2008;28:230–233. 13. Auerbach J, Ahn J, Zgonis M, et al. Streamlining the evaluation of low back pain in children. Clin Orthop Relat Res. 2008;466: 1971–1977. 14. Feldman DS, Hedden DM, Wright JG. The use of bone scan to investigate back pain in children and adolescents. J Pediatr Orthop. 2000;20:790–795. 15. Sanpera I, Beguiristain-Gurpide J. Bone scan as a screening tool in children and adolescent with back pain. J Pediatr Orthop. 2006; 26:221–225. 16. Rodriguez DP, Pousaint TY. Imaging of back pain in children. Am J Neuroradiol. 2010;31:787–802. 17. Feldman D, Straight J, Badra M, et al. Evaluation of an algorithmic approach to pediatric back pain. J Pediatr Orthop. 2006;26:353–357. 18. Shields B, Palermo T, powers J, et al. Predictors of a child’s ability to use a visual analogue scale. Child Care Health Dev. 2003;29: 281–290. 19. Ramirez N, Johnston CE, Browne RH. The prevalence of back pain in children who have idiopathic scoliosis. JBJS Am. 1997;79: 364–368. 20. Taimela S, Kujala U, Salminen J, et al. The prevalence of low back pain among children and adolescents. A nationwide, cohort-based questionnaire survey in Finland. Spine. 1997;22:1132–1136. 21. Mohseni-Bandpei M, Bagheri-Nesami M, Shayesteh-Azar M. Nonspecific low back pain in 5000 Iranian school-age children. J Pediatr Orthop. 2007;27:126–129. 22. Selbst S, Lavelle J, Soyupak S, et al. Back pain in children who present to emergency department. Clin Pediatr. 1999;38:401–406. 23. Thompson G. Back pain in children. J Bone Joint Surg. 1993; 75-A:928–938. 24. Yamaguchi K, Skagg D, Acevedo D, et al. Spondylolyis is frequently missed by MRI in adolescents with back pain. J Chil Orthop. 2012;6:237–240. 25. Cohen E, Stuecker R. Magnetic resonance imaging in diagnosis and follow-up of impending spondylolysis in children and adolescents: early treatment may prevents defects. J Pediatr Orthop B. 2005; 14:63–67.

r

2014 Lippincott Williams & Wilkins

Evaluation of a systematic approach to pediatric back pain: the utility of magnetic resonance imaging.

Several studies have suggested that back pain in the majority of pediatric patients does not have an identifiable cause. Many children undergo extensi...
140KB Sizes 0 Downloads 3 Views