Cervical Radiculopathy: Symptoms, Diagnosis, Recovery Time and Treatment Evidence

Key Takeaways

The annual age-adjusted incidence of cervical radiculopathy in a population-based study was 83.2 per 100,000, peaking at ages 50–54.[1]
C7 was the most often affected root, followed by C6; 68.4% of cases were related to spondylosis, disc, or both.[1]
In disc herniation with radiculopathy, substantial improvement tends to occur within the first 4 to 6 months.[2]
MRI changes are common in people without symptoms and must be matched to the clinical picture.[3][4]
In randomised trials, differences in disability between surgery and structured non-surgical care at 12–24 months were small or non-significant, although surgery showed advantages in some outcomes.[5][6]
In my clinical experience, the diagnosis is not always clear-cut: radicular pain can resemble a muscular neck problem or thoracic outlet syndrome, and clear weakness makes the assessment easier.
In my experience, how MRI findings are explained matters. Many degenerative changes are common with age and may be present without symptoms.[3][4] In my clinical experience, unnecessarily alarming explanations may cause patients additional worry.

Cervical radiculopathy is a common condition in which pain starts in the neck and spreads into the arm and fingers along the skin territory (dermatome) of the affected cervical nerve root.[7] On examination there may be weaker tendon reflexes, altered sensation, or muscle weakness following the same dermatomal or muscle-group pattern.[7] Neck pain without radiating arm symptoms is a broader presentation and does not by itself indicate cervical radiculopathy — it is covered in our guide to neck pain relief.

The arm pain can be intense, but the outlook is mostly favourable: in a population-based study, 90% of patients had no symptoms or only mild disability from the condition at their last follow-up.[1] Below: causes, diagnosis, recovery time, and what randomised trials show about non-surgical care, injections and surgery — plus the limited data on athletes.

How Common Is Cervical Radiculopathy, and What Causes It?

Population-based incidence figures come from a 1976–1990 survey in Rochester, Minnesota, which identified 561 patients aged 13 to 91 years.[1] The average annual age-adjusted incidence was 83.2 per 100,000 people overall — 107.3 in men and 63.5 in women — and the age-specific rate peaked at 202.9 per 100,000 in those aged 50–54.[1]

In that cohort, a confirmed disc protrusion was the cause in 21.9% of cases, while 68.4% were related to spondylosis (age-related degenerative change of the spine), disc, or both.[1] A monoradiculopathy of the C7 nerve root was the most frequent finding, followed by C6.[1]

A history of physical exertion or trauma before the symptoms began was present in only 14.8% of cases.[1] And 41% of patients had a past history of lumbar radiculopathy (nerve root pain from the lower back).[1] The median duration of symptoms before diagnosis was 15 days.[1]

In my work as a general practitioner and occupational health physician, patients with suspected or already diagnosed cervical radiculopathy come to the clinic from time to time. In some of them, it eventually turns out to be the cause of what was first assessed as neck pain. From a clinical perspective, the condition can therefore come up during the work-up of ordinary neck pain, not only as a diagnosis the patient already has.

Cervical Radiculopathy Symptoms

The core picture is neck pain radiating into the arm and fingers in a dermatomal pattern, sometimes accompanied by reduced tendon reflexes, sensory disturbance, or motor weakness in the corresponding dermatome or myotome.[7]

Pain is often severe from the outset. In a Dutch randomised trial of 205 patients whose symptoms had lasted less than one month, the mean baseline pain was 70 mm for the arm and 60 mm for the neck on a 100 mm visual analogue scale.[7] Neurological deficit in that group was mainly sensory; relatively few patients had paresis (weakness) or reduced reflexes.[7]

From a clinical perspective, pain radiating into the arm is often what first raises the suspicion of cervical radiculopathy, but it is not specific to it. Thoracic outlet syndrome can look very similar, and a stiff, painful neck can sometimes cause pain that radiates into the arm as well. In my clinical experience, the suspicion tends to be clearer when the symptoms follow a specific dermatome — for example, when the whole hand is not numb, but rather the thumb and index finger, the outer edge of the palm, or the fourth and fifth fingers, although sensory findings still need to be interpreted alongside the rest of the examination.

The opposite mistake is also possible. Radicular pain can at first look very much like a muscular neck problem, and a clinician can easily misjudge it as such. I have had cases in which the picture initially seemed muscular but later turned out to be radicular pain. For that reason, I tend to mention a small reservation to patients: even if the pain looks muscular right now, I keep the possibility of radicular pain open in case the symptoms change, persist or get worse.

How Cervical Radiculopathy Is Diagnosed

Cervical radiculopathy is primarily a clinical diagnosis. MRI of the cervical spine usually shows the underlying cause, which is most often spondylarthrosis (degenerative joint change) or a herniated disc.[7]

Physical examination tests

One commonly used provocative test is Spurling’s test: the neck is positioned — the studied versions used different combinations of extension, side-bending and rotation — and downward axial pressure is then applied to the head to see whether arm symptoms are reproduced.[8]

A 2018 systematic review in The Spine Journal identified only five diagnostic accuracy studies, and Spurling’s test was the only test evaluated in more than one of them. Its specificity ranged from 0.89 to 1.00, while its sensitivity varied from 0.38 to 0.97.[8] In practical terms, a positive test may increase the likelihood of cervical radiculopathy when it fits the history and other findings, but a negative result alone does not reliably exclude the diagnosis. The review found no studies at all that had assessed the diagnostic accuracy of the standard neurological checks — key muscle strength, tendon reflexes and sensation.[8]

In practice, I can perform Spurling’s test and check key muscle strength at the appointment, but in practice I often rely on MRI or, if the symptoms have lasted long enough, electroneuromyography to support the diagnosis. The diagnosis is not always straightforward, because many cases are not clear-cut. Clear weakness (paresis) makes the clinical assessment easier, whereas severe neck pain with radiating symptoms does not necessarily mean radiculopathy. A neurological examination for differential diagnosis is also needed, as a central nervous system cause is a possibility. In my view, pain often points the assessment towards a nerve root problem, but pain characteristics alone cannot reliably distinguish a central from a peripheral neurological cause.

Imaging: why the MRI report is not the whole story

MRI findings have to be matched to symptoms, because degenerative changes are common in people without any complaints. In a study of 63 volunteers with no history of cervical symptoms, scans were interpreted as abnormal in 19% — 14% of those under 40 and 28% of those over 40 — and disc degeneration or narrowing at one or more levels was seen in 25% of the under-40s and almost 60% of the over-40s.[4] In a larger Japanese study of 1,211 healthy volunteers, MRI showed spinal cord compression in 64 people (5.3%) and a T2 high-signal change in the spinal cord in 28 (2.3%).[3] The authors of both studies cautioned against basing operative decisions on imaging that has not been carefully correlated with clinical signs and symptoms.[3][4]

The mismatch also runs the other way. In the Dutch trial, 20–25% of the patients selected on clinical grounds had no nerve root compression on MRI.[7]

MRI also has practical limits. In my experience, if the patient’s insurance or occupational health contract does not cover it, a private MRI is fairly expensive, while access to MRI through public healthcare in Finland can take a long time. Referrals I send to public specialist care are often returned if the symptoms have not lasted for several weeks or months, unless there is clear weakness. My understanding is that this partly reflects whether a surgical assessment is likely to change the treatment. I therefore sometimes tell patients that paying for an MRI themselves may speed things up, but it is also worth asking what the scan would change: if physiotherapy is the only treatment on offer anyway, the image may add little and may not affect the treatment.

How the result is explained matters as well. In my experience, some patients become more concerned about their symptoms after learning about degenerative MRI findings. If a doctor gives the impression that the neck or back is in bad shape, patients tend to believe it, and some may conclude that the neck is ruined for good and lose hope about the future. In my clinical experience, careful and balanced explanations can help patients understand the findings without unnecessary alarm.

When I go through MRI results, I try to do it carefully. I usually explain that degenerative changes like these are very common with age, that they do not necessarily affect the patient’s life, and that having them does not mean the neck is ruined. In my experience, patients then tend to understand the situation better and are less likely to conclude that they must give up a hobby, can never exercise again, or will never be able to work again.

Cervical Radiculopathy Recovery Time: What Is the Natural Course?

A 2014 systematic review in The Spine Journal examined the course of symptomatic cervical disc herniation with radiculopathy. Of 1,221 articles identified, only three studies were accepted as having a low risk of bias.[2] Based on those, substantial improvement tended to occur within the first 4 to 6 months after onset, and time to complete recovery ranged from 24 to 36 months in approximately 83% of patients.[2] No progressive neurological deficits or myelopathy (spinal cord dysfunction) were reported in the reviewed studies — which cannot exclude such complications in other patients — and the authors stressed that the evidence was of poor quality and came from small samples in secondary or tertiary care.[2]

Randomised trial data show the early course as well. In the Dutch trial, arm pain in the wait-and-see group fell by about 3 mm per week — 19 mm in total over six weeks — while neck pain in that group did not decrease significantly during the same period.[7] After six months, most patients in all groups had no or limited pain, and the median arm pain score was zero.[7]

The longer-term population data add two qualifiers. Over a median follow-up of 4.9 years in Rochester, the condition recurred in 31.7% of patients, and 26% underwent surgery for cervical radiculopathy.[1] A combination of radicular pain and sensory deficit, as well as objective muscle weakness, predicted a decision to operate.[1]

From a clinical perspective, these patients can be difficult to help, partly because the symptoms often last a long time. In my clinical experience, patients may suffer for many months and are sometimes even desperate with the pain. Sick leave is often needed, and chronic pain can understandably add a considerable psychological burden.

Non-Surgical Treatment for Cervical Radiculopathy: What the Trials Show

Collar or physiotherapy versus waiting

The Dutch trial by Kuijper and colleagues, published in the BMJ in 2009, randomised 205 patients with cervical radiculopathy of less than one month’s duration to one of three strategies: a semi-hard collar with as much rest as possible for three to six weeks; 12 twice-weekly physiotherapy sessions with home exercises over six weeks; or continuing daily activities as much as possible without specific treatment.[7] The physiotherapy was not hands-on; it consisted of graded exercises to strengthen the superficial and deep neck muscles.[7]

Compared with the wait-and-see group, both the collar and physiotherapy groups achieved an additional 12 mm reduction in arm pain after six weeks.[7] For neck pain, compared with the wait-and-see group, the trial reported an additional reduction of 2.8 mm per week with the collar (17 mm over six weeks) and 2.4 mm per week with physiotherapy (14 mm over six weeks); neck pain in the wait-and-see group itself did not decrease significantly.[7] On the neck disability index, the collar-and-rest strategy showed a significant change relative to waiting, whereas the effect of physiotherapy was not significant.[7] At six months, pain scores in the two treatment groups did not differ from those of the control group.[7] Patients and examiners were not blinded.[7]

For a broader look at exercise-based approaches to musculoskeletal pain, see our article on isometric exercise for pain.

Epidural steroid injections

A multicentre comparative-effectiveness trial published in Anesthesiology in 2014 randomised 169 patients with cervical radicular pain of less than four years’ duration to nortriptyline and/or gabapentin plus physical therapies, up to three cervical epidural steroid injections, or a combination of both, over six months.[9] For the primary outcome — average arm pain at one month — there was no significant difference between groups (P = 0.26).[9]

On some other measures the combination group did better. Mean arm pain fell by 3.1 points on a 0–10 scale at one month with combination treatment, versus 1.8 with conservative care and 2.0 with injections (P = 0.035), and at three months 56.9% of the combination group had a positive outcome, compared with 26.8% in the conservative group and 36.7% in the injection group (P = 0.006).[9] The authors concluded that confirmatory studies are needed.[9]

In my clinical experience, many patients get only limited relief from anti-inflammatory painkillers (NSAIDs) and paracetamol. In my clinical experience, conservative physiotherapy is often started for many of these patients.

Surgery vs Non-Surgical Treatment for Cervical Radiculopathy

The operation studied in the trials below is anterior cervical decompression (or discectomy) and fusion, ACDF. Two research groups have compared it head-to-head with structured non-surgical care in randomised trials.[5][6]

Swedish trial (Engquist et al., Spine, 2013). Sixty-three patients were randomised to ACDF followed by physiotherapy (n = 31) or the same structured physiotherapy programme alone (n = 32) and followed for 24 months.[6] There was no significant between-group difference in the neck disability index. Neck pain intensity favoured the surgical group, which improved more rapidly during the first postoperative year, but the differences between groups decreased after two years.[6] The authors concluded that structured physiotherapy should be tried before surgery is chosen.[6] A companion analysis found no significant differences in neck range of motion, neck muscle endurance, or hand function (P = 0.17–0.91).[10]

Norwegian trials (Taso et al., NEJM Evidence, 2025). Two parallel randomised trials at Oslo University Hospital enrolled 180 patients with disabling radicular arm pain — 89 with cervical disc herniation and 91 with spondylosis — and compared ACDF with non-surgical treatment consisting of three sessions with physical medicine and rehabilitation physicians and three sessions with physiotherapists for functional and cognitive-behavioural support.[5] At 12 months, the disc herniation trial showed a statistically significant difference in the neck disability index in favour of surgery (7.4 points; 95% CI 1.6 to 13.3), whereas the spondylosis trial showed no difference (2.3 points; 95% CI −4.9 to 9.6).[5] For context, the authors set the minimal important difference on this 0–100 scale at 15 points.[5] Two non-surgical patients in the disc herniation trial and 11 in the spondylosis trial crossed over to surgery.[5]

Taken together, the between-group differences in disability at 12–24 months were small or statistically non-significant. The Norwegian disc-herniation trial nevertheless found a statistically significant benefit of surgery, smaller than its predefined 15-point minimal important difference, and in the Swedish trial neck pain improved more in the surgical group.[5][6]

Fairly often, I refer these patients to a cervical spine surgeon — a neurosurgeon or a spine orthopaedist — who assesses and operates on these conditions. If the assessment does not support surgical treatment, patients are usually referred on to a physiatrist (a specialist in physical and rehabilitation medicine).

Cervical Radiculopathy in Athletes

Evidence on return to sport after cervical spine surgery is sparse and largely retrospective, and it does not cover athletes with cervical radiculopathy who are treated non-surgically. A 2016 systematic review in Global Spine Journal identified nine observational, retrospective case series with 175 patients who had undergone cervical spine surgery.[11] Among professional athletes, 75% (76/102) returned to their sport after surgery for mostly herniated discs; among recreational athletes aged 10 to 42 years, 76% (51/67) returned.[11] The review also included six snowboarders operated on for cervical fractures, none of whom returned to snowboarding.[11] The authors concluded that successful return to competitive sport is well described after single-level ACDF for disc herniation, while return-to-play outcomes for other cervical diagnoses and procedures remain unclear.[11]

The review excluded athletes who were not treated surgically, so it cannot show whether surgery gives better return-to-play rates than non-surgical care.[11] In the Rochester population data, only 14.8% of cases were preceded by physical exertion or trauma.[1]

What the Evidence Means in Practice

Diagnosis is clinical first; MRI mainly shows the cause.[7]
Spurling’s test has shown high specificity but variable sensitivity (0.38–0.97), so it is interpreted alongside other findings.[8]
Scan findings are common without symptoms, including cord compression in 5.3% of healthy volunteers.[3]
A collar or physiotherapy sped up early pain relief but did not change the six-month result in the Dutch trial.[7]
Recurrence is common: 31.7% over a median 4.9 years.[1]
Manual therapy was not part of the hands-off physiotherapy in the Dutch trial, and none of these trials tested it as a separate treatment; for evidence on spinal manipulation, see our article on chiropractic for athletes.

In practice, the warning symptoms I go through with patients include weakness or paralysis developing in the arm and clumsiness of the hand. In my view, such symptoms may warrant prompt medical assessment.

Conclusion: Cervical Radiculopathy Usually Improves, but Slowly

Cervical radiculopathy can cause substantial pain, but many patients improve over time. Substantial improvement in disc herniation tends to come within the first 4 to 6 months.[2] A collar or physiotherapy sped up early pain relief without changing the six-month outcome,[7] and in randomised trials surgery showed small or non-significant differences in disability at 12–24 months compared with structured non-surgical care, with advantages in some outcomes.[5][6] Recurrence in 31.7% of patients over a median 4.9 years shows that a later episode is a real possibility.[1]

In my clinical experience, the course is often long and can bring sick leave and a considerable psychological burden, and careful explanation of MRI findings is an important part of helping patients understand their condition.

Cervical Radiculopathy: Symptoms, Diagnosis, Recovery Time and Treatment Evidence

Table of Contents

Key Takeaways

How Common Is Cervical Radiculopathy, and What Causes It?

Cervical Radiculopathy Symptoms