Intended for healthcare professionals

Practice Clinical updates

Hip osteoarthritis

BMJ 2016; 354 doi: (Published 06 July 2016) Cite this as: BMJ 2016;354:i3405
  1. Nick Aresti, specialist registrar in trauma and orthopaedic surgery and National Medical Director’s clinical fellow1,
  2. Jamila Kassam, lead orthopaedic research physiotherapist1,
  3. Nick Nicholas, patient with hip osteoarthritis and consultant gynaecologist2,
  4. Pramod Achan, clinical director and consultant orthopaedic surgeon1
  1. 1Barts Health Trauma and Orthopaedics, Royal London Hospital, London E11BB, UK
  2. 2The Hillingdon Hospitals NHS Foundation Trust, London, UK
  1. Correspondence to: N Aresti email{at}

What you need to know

  • Hip osteoarthritis is a huge burden on the NHS and affects almost 2.5 million people in England

  • The main investigation used to diagnose hip osteoarthritis is plain radiography

  • Weight loss, activity modification, and physiotherapy form the mainstay of non-pharmacological treatment

  • Paracetamol and non-steroidal anti-inflammatory drugs (NSAIDs) are the most commonly used analgesics

  • Total hip replacement is a successful surgical option in the treatment of osteoarthritis

Osteoarthritis is a degenerative joint disease that causes progressive damage to articular cartilage and surrounding structures. The hip is the second most commonly affected joint (after the knee), and around 11% (2.46 million) of people in England are affected.1 This review updates readers on current management options for hip osteoarthritis.

Sources and selection criteria

We searched PubMed and Google using the terms “hip OA” and “hip osteoarthritis.” We used NICE and OASRI guidelines to provide up to date recommendations on treatment options, and independently evaluated references from the guidelines.

Who gets osteoarthritis of the hip?

Risk factors can be divided into general, intrinsic, and extrinsic. General factors include age, sex, and genetics. Hip osteoarthritis is almost twice as common in women than in men,1 2 and genetic studies show a 50% heritability caused by European genetic variants. Intrinsic factors include incongruency (such as dysplasias) and joint laxity: they cause abnormal wear and loading, and therefore accelerate articular degeneration. Extrinsic factors such as increasing body mass index, high levels of certain exercise, and heavy manual labour are thought to increase the incidence and progression of hip osteoarthritis.3 4 5

Most often the cause of hip osteoarthritis is multifactorial.6 A series of risk factors lead to instability, malalignment, increased joint load, microtrauma and structural damage. The joint responds through subchondral and synovial inflammation, and bone hypertrophy. This is visible on radiographs as narrowed joint space, sclerosis, and cyst or osteophyte formation.7

Hip osteoarthritis is associated with other diseases, but there is often no proven causal relationship. A population based cohort study showed that hip osteoarthritis is associated with frailty, with an odds ratio after adjustment for confounding variables of 1.57 (95% confidence interval 1.1 to 2.22).8 Hip osteoarthritis is also associated with an increased risk of all-cause mortality (hazard ratio 1.14)9 and higher rates of mental health problems.10 One large, population based cohort study also suggests an increase in cardiovascular mortality associated with osteoarthritis (hazard ratio 1.24).9 A prospective, population based cohort study11 suggests this is probably because of ensuing disability rather than the presence of osteoarthritis itself.

What are the symptoms and clinical signs of hip osteoarthritis?

Patients with hip osteoarthritis typically develop pain over months to years rather than acutely. They describe an achy groin pain, which is at first intermittent, worse at the end of the day, and activity related (particularly while walking or climbing stairs). Rest and night pain may feature in all disease stages, but severity increases as the disease progresses.12 Night pain, particularly when atypical or out of context of other osteoarthritis symptoms, may be a sign of a more serious diagnosis (see table 1 for examples). Similar hip pain can occur in other conditions, some of which are highlighted in table 2. Cases of hip osteoarthritis may be considered “severe” if symptoms are a substantial barrier to mobility and independence. Several scoring systems score the severity of symptoms. The Oxford Hip Score is perhaps the most commonly used and forms part of the Patient Reported Outcome Measures (PROMs) in England for total hip replacements.13

Table 1

“Red flags” and risk factors for diagnoses other than osteoarthritis

View this table:
Table 2

Differential diagnoses of hip pain

View this table:

Typical examination findings include groin tenderness and reduced or painful movements.14 15 Internal rotation is the most sensitive single indicator of hip osteoarthritis.15

How do I diagnose hip osteoarthritis?

If a patient presents with typical features of hip osteoarthritis, clinical evaluation alone is sufficient to rule out other diagnoses: further investigation is not necessarily required.16 The National Institute for Health and Care Excellence (NICE) suggests that osteoarthritis can be diagnosed “clinically without investigations” if a patient is aged ≥45 years and has activity related joint pain and either no or minimal morning joint related stiffness .16 If there is uncertainty, symptoms worsen, or red flag or atypical symptoms are present, consider referral for specialist opinion.


Plain radiography is the simplest, least expensive, and most commonly used method of assessing hip osteoarthritis.17 It cannot visualise cartilage or soft tissue, but joint space provides an estimate of the severity of cartilage damage. Plain radiography has been shown to have high specificity (0.76-0.90) and variable sensitivity (0.44-0.78) compared with magnetic resonance imaging (MRI) in detecting hip osteoarthritis.18 Plain radiography should be used for diagnostic purposes unless there is uncertainty (such as a symptomatic patient but limited radiographic evidence) or soft tissue disorders are suspected.19

Typical radiographic features are highlighted in figure 1. The Framingham Osteoarthritis Study, a community based study of osteoarthritis, found that 15.6% of patients with frequent hip pain had radiographic evidence of osteoarthritis, while only 20.7% of patients with radiographic hip osteoarthritis had frequent pain.20 This suggests that many people with pain will not have radiographic changes, and, conversely, many people with radiographic osteoarthritis are not symptomatic. When patients fall into either category, their symptoms should be monitored while managing them conservatively.


Fig 1 Typical plain radiograph of bilateral hip osteoarthritis Notice the loss of joint space (1), osteophytes (2), subchondral sclerosis (3), and subchondral cysts (4). In this advanced case, there is also evidence of deformity of the femoral head (5). Joint space narrowing is the best diagnostic criterion7

What are the treatment options?

Patients often see osteoarthritis as a function of age rather than a medical disorder, and younger patients are often more distressed and frustrated with managing the disease.21 22 When deciding on treatment options with a patient, consider social and psychological factors, such as function, quality of life, occupation, mood, relationships, and leisure activities.16 Shared decision making tools aim to provide patients with a greater degree of information on possible treatments, reducing decisional conflict, while improving decision quality and patient engagement. Randomised controlled trials have shown shared decision making tools to be beneficial to both patients and clinicians, and their use is encouraged.23 A variety of decision aids are available via the NHS website.24

Osteoarthritis Research Society International (OASRI) recommends a combination of both pharmacological and non-pharmacological methods to treat hip osteoarthritis.25 The mainstay of surgical treatment is total hip replacement.

Non-pharmacological methods

Weight loss

Several studies demonstrate improved function and a reduction in disability after weight loss in patients with knee osteoarthritis.26 27 28 A meta-analysis of 35 trials suggests weight loss of >5% is associated with a significant reduction in patient self reported disability due to knee pain.28 There is less robust evidence of improved function with weight loss for hip osteoarthritis.29 Nevertheless, expert consensus recommends weight loss in patients with hip osteoarthritis,16 25 through a reduction in calorie intake, enrolment in weight loss organisations, and non-joint loading exercises such as swimming.24

Physical therapy

Increasing muscle strength improves the mechanical environment and reduces joint loading of an arthritic hip. A Cochrane review found that completion of a supervised physiotherapy programme reduces pain and improves physical function in patients with mild to moderate pain from hip osteoarthritis. The benefits of supervised physiotherapy programmes are small but are shown to last three to six months after treatment.30 OASRI and NICE recommend that all patients with hip osteoarthritis are offered physiotherapy, based on analysis of current literature, guidelines, and expert consensus.16 25 31 Physiotherapy should include local strengthening and general aerobic exercises.16 A combination of class based and individual home based physiotherapy programmes are thought to be cost effective and to improve patient outcomes.32

Other treatment options

NICE recommends a variety of adjuncts to the core treatments outlined above. These include electrotherapy (such as TENS machines) and thermotherapy (such as heat or cold pads). Acupuncture is not recommended, after a detailed appraisal of evidence showed a lack of benefit compared with sham acupuncture.16

Pharmacological methods

A variety of analgesics, including paracetamol, NSAIDs, and opioids are used to manage pain.

A Cochrane review looked at 15 trials that evaluated the use of paracetamol versus placebo and NSAID in treating hip osteoarthritis. Compared with placebo, paracetamol led to only a small reduction in pain (standardised mean differences 0.13 (95% CI 0.22 to 0.04)). NSAIDs were moderately superior to paracetamol in pain reduction, physician global assessments, and functional status. The superiority of NSAIDs was more marked in severe osteoarthritis. NSAID groups had a higher rate of gastrointestinal events (relative risk 1.47), but otherwise there was no significant difference in safety between paracetamol, placebo, and NSAIDs.33

Prescribers should consider patient preferences, clinical judgment, cost, and safety when prescribing anaglesics.33 NICE recommends a stepwise approach to pharmacological management, with paracetamol as first line treatment, initially guided by pain but with scope to increase doses as required. Should this prove insufficient, it should be substituted with an oral NSAID plus a low acquisition cost proton pump inhibitor.16 Patients should be kept on the smallest required dose of NSAID for the shortest time possible, and monitored for gastrointestinal, liver, and cardiorenal side effects.

If paracetamol or NSAIDS are insufficient for pain relief, then consider adding an opioid such as tramadol.16 A meta-analysis of 18 randomised, placebo controlled trials assessed the efficacy of various opioids for osteoarthritis. The pooled effect sizes of all opioids against placebo for pain intensity and physical function were 0.79 (95% CI 0.98 to 0.59) and 0.31 (0.39 to 0.24) respectively.34 Stronger opioids (such as oxycodone) have a greater reduction in pain intensity but a worse side effect profile. Weaker opioids are recommended for refractory cases during periods of increased pain by NICE and OARSI.16 25 Opioids should be prescribed with caution because of their side effect profiles and risk of dependency, particularly in elderly patients.16 If regular use is required, referral to a pain clinic or an orthopaedic surgeon should be considered.

Intra-articular injections

Consider intra-articular injections for the acute relief of pain not responding to oral analgesia, for acute inflammation, or where the injection can be used to confirm a diagnosis of osteoarthritis.25 Corticosteroids have potent anti-inflammatory properties and reduce the degree of synovitis in osteoarthritis. Although the relation between active synovitis and pain is unclear, it is thought that inflammation plays some role in modulating pain.16 Randomised controlled trials demonstrate a short term benefit in pain and function in arthritic hips after a corticosteroid injection,16 35 36 37 but intra-articular injections are technically challenging, require an image intensifier or ultrasound scanning, and often need the patient to be sedated. Patients may be referred directly for injections or via an orthopaedic surgeon.


Total hip replacement remains one of the most commonly performed and successful operations in the world.38 Total hip replacement involves removing the articular surfaces of the joint and replacing them with prostheses.

There is no clear evidence on when to refer for surgery, and existing guidelines are based only on expert consensus.16 25 NICE recommends that patients are considered for surgery if their symptoms have a substantial impact on their quality of life and they have a poor response to non-surgical treatment.16

Patient-specific factors should not be a barrier for referral. Each decision remains unique, and patients must calculate their own risk-benefit ratio based on symptom severity, general health, expectations, and treatment effectiveness.16 Postoperative and prosthesis related complications are increased in smokers and the obese. A cohort study found a greater risk of systemic postoperative complications for obese patients (odds ratio 1.58 (95% CI 1.06 to 2.35)) and for current smokers (odds ratio 1.56 (1.14 to 2.14)) compared with non-obese, non-smoking patients.39 A separate meta-analysis showed a relative risk of total hip replacement revision of 2.58 and of deep infection of 3.71 for smokers.40 Potential risks after total hip replacement are summarised in tables 3 and 4.

Table 3

Potential risks of major intraoperative complications after primary total hip replacement

View this table:
Table 4

Reasons for revision of total hip replacements as a proportion of total number revised or of primary operations performed between 2003 and 2014 (from National Joint Registry41)

View this table:

An important determinant of the success of a total hip replacement is how long it lasts, and the only real measurable endpoint is whether it is revised. National Joint Registry data suggests the probability of needing a total hip replacement revision at 11 years is 6.20%.41 Around half of revisions occur as a result of “aseptic loosening” of the prosthesis, the main symptom of which is thigh or groin pain.41 Radiographs readily diagnose the issue. Aseptic loosening can be due to a variety of reasons, including patient related factors (such as body mass index and activity level), surgical technique, and prosthesis design. Age is an important determinant of the risk of revision, as shown by the National Joint Registry41 and a cohort study of 4668 patients undergoing primary hip and knee replacements, whose findings are summarised in table 5.42

Table 5

Risks of revision of hip and knee replacements by age (from a New Zealand patient cohort study42)

View this table:

A patient’s perspective

I didn’t realise I had osteoarthritis since only one hip joint was painful. Although I was overweight, I considered myself relatively fit. I was 60 years old.

My first symptoms were pain and restriction of movements. I couldn’t sit cross-legged without pain, and, as the osteoarthritis got worse, I began to limp.

I initially took regular simple analgesics, which provided temporary relief lasting only a few hours. I developed coping mechanisms to avoid pain, by sitting down wherever possible and I avoided of any strenuous activity.

Once diagnosed with osteoarthritis, I continued with my analgesic cocktail but also had an injection of steroid and local anaesthetic. The procedure was quick and relatively painless. Within 24 hours I noticed a significant improvement in my pain. However, this only lasted two weeks, after which I was back to square one.

I didn’t want to rush into major surgery too soon, and wanted to optimise the best time to do it. My orthopaedic surgeon told me that I would know when I was ready for surgery when I couldn’t put my socks on when getting up in the mornings.

I plodded on for a few more months, and eventually one morning, when getting dressed to go to work, I had to wake up my wife to help me get my socks on. I was ready for surgery.

My surgery was successfully performed under spinal anaesthetic. I stayed in hospital for two days and was allowed home with simple analgesia and oral anticoagulant. I was also advised to wear compression stockings. This was the hardest part of my recovery. The stockings were tight and soon became itchy and claustrophobic. My wife had to struggle to change them for me whenever I took a shower. The other problem I had was sleeping on my back. As someone who never did, I struggled to sleep and was worried that I would cross my legs inadvertently while asleep and dislocate my hip. Eventually I slept on my side with a pillow between my knees.

I did my exercises religiously and within three weeks of surgery I was walking unaided without a stick. I stopped taking any analgesia after about two weeks post-operation.

Within six weeks I was back to work doing my job unhindered by pain or restriction of movement.

Within two years I had to repeat the process to have my second hip done. My recovery was quicker, and all went well. It was all déjà vu, and I knew what to do.

Since the surgery, I enjoy walking without pain and I can sit cross-legged on the floor. Soon after surgery I had an episode of low back ache, which required physiotherapy and acupuncture. It lasted six months and was probably due to the change in my posture while walking. I think patients having hip surgery ought to be aware that this might happen.

I am glad I had the operation and fully understand and appreciate why it can make a difference to someone’s life. It transformed me from being almost housebound to fully mobile and integrated back into society.

Patient involvement

Patients were recruited to plan, write, and review the article. A patient was also asked to describe his personal experience with hip osteoarthritis and its treatment.


  • Competing interests: We have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; externally peer reviewed.


View Abstract

Log in

Log in through your institution


* For online subscription