Detecting Developmental Dysplasia of the Hip (DDH)
Lee Segal, MD
- Explain the controversies surrounding DDH screening
- Reference criteria developed to determine if a particular condition is worthy of screening
- Describe the importance of the physical exam in DDH screening
- Determine when it is appropriate to obtain hip US studies and when to refer to pediatric orthopedics.
A 10-day-old female infant presented to the pediatric orthopaedic clinic for evaluation of possible hip dysplasia. The parents reported a history of a left hip click at birth. She was a double footling breech delivered by caesarian section at 39 weeks. She was a first-born child to a family without a family history of DDH. The family was instructed on double diapering the child, and safe swaddling. The hips were stable on exam at this visit and at a subsequent clinic visit at four weeks of age, with negative Ortolani and Barlow tests. A hip ultrasound performed at six weeks of age was normal. The left hip had an alpha angle of 60°, had coverage greater than 50% (normal: alpha angle ≥ 60° and > 50% coverage), and was stable with stress maneuvers (Figure 1).
An infant presenting to the primary care provider with the concern of Developmental Dysplasia of the Hip (DDH) like the case presented continues to be problematic. When are additional imaging studies such as a hip ultrasound indicated? When is it appropriate to refer a child to an orthopaedic surgeon? Published guidelines to address these questions and other DDH related issues, such as screening, have not been consistent. The United States Preventive Services Task Force (USPSTF) 2006 report stated that there was insufficient evidence to support DDH screening. Earlier reports from the American Academy of Pediatrics (AAP) and other professional organizations strongly supported the role of DDH screening.
The confluence of many factors has fueled this controversy. These include concerns of over-treatment and potential adverse outcomes, an emphasis on data-driven analysis, and an ill-defined relationship between clinical and imaging risk factors and the natural history of DDH.
The goal of screening programs is early detection and prevention of future disability from medical diseases. The US Commission on Chronic Illness in 1951 defined screening as the presumptive diagnosis of unrecognized disease or defects by the application of tests, examinations, or other procedures, which can be applied rapidly. In that context nearly 75 years ago, the most common screening tests were blood glucose determination, a serological test for syphilis, radiography for chest pathology, and cytology for cancer detection. Early application of screening tests was performed during military conflicts to screen for psychiatric disease (WWI) and syphilis (WWII). The era of preventive healthcare following WWII introduced numerous screening programs such as the Pap smear (1947), school screening for scoliosis (1962), phenylketonuria (1961), and mammography (1970). It was at the latter end of this era that several reports provided scientific rationale and guidelines for screening medical conditions as a public health measure. Wilson and Jungner described ten criteria to consider when determining the utility of screening for a disease (Table 1). Frankenburg discussed the validity, sensitivity, and specificity of screening tests; and how the prevalence of a disease impacts the predictive value of a test (lower prevalence results in a higher percentage of false positive findings, and the subsequent cost and resource implications). Ethical concerns of screening have also been raised. Screening changes the traditional physician-patient relationship, as the medical establishment seeks out the “asymptomatic patient.”
Table 1. Wilson and Jungner classic DDH screening criteria
Wilson JMG, Jungner G. Principles and Practice of Screening for Disease. WHO, 1968
Before critically evaluating DDH as a medical condition worthy of screening, the evidence is clear that early detection and treatment of DDH prevents the development of future disability such as premature osteoarthritis and possible need for hip replacement. DDH is asymptomatic during infancy and early childhood and therefore screening of otherwise healthy infants is performed to detect this uncommon condition.
The true incidence for the entire spectrum of DDH is not known. The incidence of newborns with hip instability has been reported to be 1 in 100. True dislocation of the hip in newborns is thought to be 1 in 1000. The difficulties defining true incidence are influenced by the age of the child at the time of examination, the population being studied, the skill and experience of the examiner, and the diagnostic criteria applied. The cornerstone for DDH screening is the physical examination. The 2000 AAP Clinical Practice Guideline recommended that all newborns be screened for DDH by physical examination and with follow-up at scheduled well-baby periodic examinations.
Several studies have emphasized the importance of training and experience in clinical screening for DDH. Wirth et al. noted that the quality of the clinical exam has always been the key to successful clinical screening for DDH and remains the most important part of any selective screening program for “at risk” babies. Training in the examination for neonatal hip instability (NHI) as part of DDH screening is critical to the success of selective hip US screening strategies. Price et al noted that the accuracy of the clinical exam improved from (13-24%) to (53-60%) with appropriate teaching and experience. Concerns have been raised that musculoskeletal (MSK) training in primary care residency programs and postgraduate education has not received the proportionate attention that DDH and other MSK conditions warrant. Despite the introduction of screening programs, the incidence of late onset hip dislocation (1 in 5000) persists.
In 1980, hip ultrasonography (US) was introduced as an adjunct to screening and treatment for DDH. Hip US provided a method to image the hip joint that is predominantly cartilage prior to 4-6 months of age (Figure 2). Hip US in infants has improved our understanding of DDH, that this disorder represents a wide spectrum of hip dysplasia. No longer are hips either normal or dislocated. Rather, a large “gray zone” of sonographic abnormalities has been defined. The intended goal of hip US, as an adjunct to the physical exam was to decrease the incidence of late presentation DDH. With any new imaging technology, such as magnetic resonance imaging (MRI) when it was first introduced, the diagnostic information provided by hip US often exceeds our understanding of the disease. Has hip US been successful in achieving these goals? Debate continues if universal or selective hip US screening should be used, or if static versus dynamic screening, or a combination of both methods is more effective. For the practitioner in a rural area, access to an imaging center that does a high volume of hip US studies to refer patients is not always available. Even in locations with access, has the hip US study become a “default” or a substitute for the physical exam? Future research is needed to better define what hip US abnormalities in infancy correlate with true hip dysplasia. Even among experts, consensus on methods, classifications, and sonographic abnormalities in the immature hip is lacking.
Figure 2. Infant hip joint is predominantly cartilage
DDH Risk Factors
An important part of screening is the identification of risk factors associated with specific disorders such as DDH. The infant described in our case was found by prenatal ultrasound to be in a breech position, and was delivered by C-section. Time honored risk factors (the 4 “F’s”) for DDH include breech position, female, first born, and positive family history. Torticollis, foot deformities (metatarsus adductus / calcaneovalgus) and oligohydramnios have also been considered as clinical risk factors for DDH. Hip clicks, opposed to a true hip “clunk” as described by Ortolani with hip reduction, are not consider a risk factor for DDH. A recent survey showed poor consensus of these risk factors for DDH from a group of experts. In the United Kingdom, a strong family history and breech presentation are considered “true” risk factors as defined by the NHS/Standing Medical Advisory Committee.
It is common that a risk factor of etiological importance be considered as a “screening test” for a disorder. However, it is not widely recognized that a risk factor must be strongly associated with a disorder to be a meaningful screening test. The strength of association between a risk factor and a disorder can be quantified by the relative risk or relative odds ratio for continuous data. Wald and co-authors noted that a relative odds ratio of 5 between the highest and lowest fifth of the distribution of a risk factor (ROQ1-5) results in only a 14% detection rate for a 5% false positive rate if the standard deviations of the risk factor in people with and without the disorder are equivalent. As an example, the ROQ1-5 for maternal serum α fetoprotein and open spina bifida at 16 to18 weeks gestation is 246, that 91% of affected pregnancies can be detected with a false positive rate of 5%.
The clinical practice guideline for the early detection of DDH from the AAP in 2000 presented relative and absolute risk factor data. Relative risk (RR) provides context, comparing various risk factors for a medical condition. Conversely, absolute risk lacks context, but better defines actual risk. In their model driven analysis, the AAP report determined relative and absolute risks for gender, family history (1st degree relative), and breech. The relative risk of DDH without risk factors as a baseline was 4.6 for girls, and 1.0 for boys. The RR with a positive family history was 1.7, and for breech 6.3 (no differentiation of types). The risk factors are thought to be additive. The absolute risk per 1000 live births was 32 (for female / + FH); and 133 (for female / breech).
As noted above, breech presentation may be the most important single risk factor, with DDH reported in 2 to 27% of males and females presenting in the breech position. Frank breech in a female (sacral presentation with hips flexed and knees extended) appears to have the highest risk. Most evidence supports breech position toward the end of pregnancy rather than breech delivery that contributes to DDH. Mode of delivery (i.e., Caesarian section) may decrease the risk of DDH with breech positioning.
In general, risk factors are poor predictors of DDH. Being female alone without other known risk factors accounts for 75% of DDH. For the practitioner, recommended strategies to determine the need to selectively obtain a hip US study in an infant with associated risk factors and a stable hip exam are still evolving. From a medico-legal perspective, have the risks of missing a late presentation of DDH resulted in the overuse of additional imaging studies (hip US or radiographs after four months of age), and potential overtreatment? This emphasizes the importance of a careful physical examination of all infants in detecting DDH.
The USPSTF Report on DDH Screening
The USPSTF is composed of a panel of national experts in evidenced based medicine and prevention that provide evidence-based recommendations about clinical preventive services such as screening. The task force assigns a letter grade (A, B, C, D, or I) for defining and suggestions for practice. In a 2004 report, the USPSTF gave adolescent idiopathic scoliosis screening a D rating, recommending against this service. In a 2006 report, An “I” grade was given to DDH screening in the 2006 report by the USPSTF. An “I” rating denotes “that the current evidence is insufficient to assess the balance of benefits and harms of the service. Evidence is lacking, of poor quality, or conflicting, and the balance of benefits and harms cannot be determined.”
Unfortunately, the 2006 USPSTF report giving an “I” or insufficient grade leaves practitioners without a compass to direct decision-making about screening in the care of infants presenting with DDH. Former panel members of the USPSTF noted that grades were heavily weighted toward evidenced based medicine (EBM) and predictors of poor health (PPH).
The report cited poor evidence of the effectiveness of both surgical and non-surgical interventions. Wyer et al. provided a conceptual history of EBM with its roots in clinical epidemiology, and how the hierarchy of evidence has become an integral part of the medical peer-review landscape. Data driven systematic review models tend to equate the lack of quality studies to the lack of evidence. David Sackett defined EBM as the “integration of best evidence with our clinical expertise and our patient’s unique values and circumstances.” In the clinical context of caring for infants, would a clinic practice model that takes into account the contributions of patient’s values, understanding pathophysiology of disease, and a physician’s clinical experience such as the one used in the 2000 AAP clinical practice guideline be more appropriate?
The USPSTF places an increased emphasis on the concept of PPH, and the 2006 report identified avascular necrosis (AVN) as a primary PPH. The USPSTF noted that AVN is the most common and severe cause of harm of both surgical and non-surgical interventions. These rates may be overstated, as more recent studies describe the risk of AVN as less than 1% with screening, early detection, and the use of a abduction splinting (e.g. Pavlik harness).
Should we dismiss high quality evidence that supports the value of DDH screening in reducing the rate of late DDH presentation? Two random clinical trial studies by Rosendahl and Holen demonstrated that both selective and universal screening reduce the prevalence of late presentation of hip dysplasia (0.13-0.7/1000). The UK Hip Trial found that infants in the US hip exam group were less likely to have hip abduction splinting without increasing the need for surgery in the first 2 years. In their clinica lexperience with selective hip US, Clarke et al. noted a decrease in the late presentation of DDH (0.34/1000), and a decrease in infants requiring surgery (0.74 in 1000). Decision analysis models have been an effective complementary approach to support DDH screening.
Is DDH a medical condition worthy of screening?
The application of the Wilson and Junger screening criteria (Table 1) is a useful method to determine if DDH is worthy of screening as a public health measure. Although open to debate, the current literature suggests that eight of the ten screening criteria are fulfilled. We believe that DDH is a condition of value to screen. The condition may initially be occult, it is easy to treat if detected early, and more difficult to manage when discovered late. A reasonable goal is to detect a subluxated or dislocated hip by six months of age.
The exam in the neonatal period (up to three months of age) is thought to be acceptable. The exam includes observing for a limb length discrepancy, asymmetric thigh or gluteal skin folds, and performing the Ortolani and Barlow tests. Although described independently, these two neonatal provocative tests are performed together in a continuous manner. Starting with the hips flexed and adducted, with gentle pressure on the greater trochanter followed by gently abducting the hip (moving away from the midline), the examiner can feel whether or not the femoral head reduces into the acetabulum over hypertrophied articular cartilage. This is the “clunk” described by Ortolani as the posterior dislocated hip is positioned or reduced (Figure 3). This should not be confused with high pitched hip clicks often elicited with hip flexion and extension. These have been proven to be inconsequential, but differentiating true hip dislocation (clunks) from hip instability, and from benign hip clicks requires experience in the practice setting. The examiner should never forcibly attempt to dislocate the femoral head with the Barlow test (moving the hip toward the midline with downward pressure). The majority of positive Barlow hips resolve on their own. Forceful and repeated Barlow maneuvers have the potential risk of stretching the hip capsular tissue and breaking the airtight seal of the hip joint. The critical question is to determine if the hip is dislocated and can be reduced, not to prove that the hip can be dislocated. After three months of age, asymmetric hip abduction becomes the most important sign of DDH. Bilateral hip dislocations or less common conditions such as congenital coxa vara can be diagnostic challenges.
Figure 3. Ortolani test
The natural history of the disease and an agreed upon policy on who to treat are the two outstanding criteria that warrant further research. The natural history of untreated DDH is poorly understood, and predicting the outcome of neonatal hip instability (NHI) is not possible. This is particularly true to Graf type II hips noted on hip US in which the degree of hip dysplasia (depth and shape of the hip socket or acetabulum) may be due to physiologic immaturity or mild dysplasia. There is universal agreement that Ortolani positive hips (dislocated hips that reduce with hip abduction) or dislocated hips should be treated with some form of abduction splinting like Pavlik harness treatment. It is with the other types of NHI where no treatment consensus has been established. Refinements in hip ultrasound (techniques, screening strategies, and classification systems) that better define which infants with NHI are at risk for adverse outcomes such as early osteoarthritis will be important to satisfy these two criteria for screening.
Nomenclature for DDH Detection
The concept of screening to discover a medical condition and prevent future disability with early detection has evolved over time. The tests were not intended to be diagnostic, but advances in medicine have made many tests such as those found on newborn screening panels to be a one time exam, and no longer a continuous process. DDH does not easily fit into this binary concept. In fact, the nomenclature for DDH itself changed from congenital dislocation of the hip (CDH) to developmental dysplasia of the hip (DDH) to reflect not only the wide spectrum of the disease, but also the evolving dynamic nature of the condition. Despite our understanding and appreciation that the clinical examination of the infant hip remains the “cornerstone” of screening for DDH, selective hip US screening in North America has become a substitute or default for the physical exam. This in part is due to the medical legal concerns, to lack of proper MSK training, to the lack of consensus of risk factors, and to the lack of understanding the natural history of NHI. Too often, hip US is not used as an adjunct to the physical exam to confirm DDH but an imaging modality that may be over-utilized, improperly ordered, and may lead to over-treatment.
We support the proposal as suggested by others to change the nomenclature of DDH detection from screening to surveillance. This would emphasize the importance of the physical exam, differentiate the physical exam from US studies as a screening tool, and encourage DDH evaluation as a continuous process. This would be in alignment with the AAP definition of surveillance, which is the longitudinal continuous process where by health care professionals identify children who may have developmental problems (Pediatrics 2006; 118(1): 405-420).
- Despite the inconclusive grade from the 2006 USPSTF report, other professional organizations such as the AAP, POSNA, and the AAOS continue to recommend newborn and periodic physical exam screening for DDH.
- Treatment of neonatal DDH is not an emergency. In-hospital initiation of bracing by the primary care physician or other health professional is not required as most newborn positive Barlow tests will resolve spontaneously. However, infants with persistent instability should be referred to an orthopaedic surgeon within several weeks. Initiation of treatment is based on the clinical examination of hip instability.
- Tight swaddling of the lower extremities with the hips extended should be avoided. The concept of safe swaddling, which does not restrict hip motion, minimizes the risk of DDH.
- Evidence strongly supports screening for and treatment of hip subluxation and dislocation (Ortolani positive) and observation of milder early forms of dysplasia and instability.
- It is acceptable to refer children with suspected DDH or with positive risk factors to a pediatric orthopaedist without a prior US, which is preferable to obtaining an improperly timed or poorly performed study.
- Hip US can be useful between ages 4 weeks and 4 to 6 months for high-risk infants. “High-risk” is a relative and practitioner-dependent term, but considerations are first-born female, breech presentation, positive family history, parental concern, suspicious but inconclusive periodic exam after age 4 weeks, and others per CPG and practitioner comfort. Proper timing, performance and interpretation of infantile hip US per AIUM and ACR guidelines is critical to avoid under or overtreatment.
- Radiographs (AP and frog pelvis views) are useful after age 4 to 6 months for the high-risk baby or positive clinical findings. 8. A reasonable goal for the primary care physician should be to prevent hip subluxation or dislocation by six months of age using the periodic examination. Selective hip US or radiography should be used in consultation with the pediatric radiologist and orthopedist.
- Wilson JM, Jungner YG. Principles and Practice of Screening for Disease. WHO. 1968
- Frankenburg WK. Selection of diseases and tests in Pediatric screening. Pediatrics 1974;54;612-616
- Shipman SA, Helfand M, Moyer VA, Yawn BP. Screening for developmental dysplasia of the hip: a systematic literature review for the U.S. Preventive Services Task Force. Pediatrics. 2006; 117: e557–76. Recommendation Statement.
- Wyer PC, Silva SA. Where is the wisdom? I - a conceptual history of evidence-based medicine J Eval Clin Pract 2009; 15:891-898.
- Clinical practice guideline: early detection of developmental dysplasia of the hip. Committee on Quality Improvement, Subcommittee on Developmental Dysplasia of the Hip. American Academy of Pediatrics. Pediatrics 2000; 105: 896-905.