When a child persistently complains of vague knee pain, they may end up in a sports therapy clinic. Patrick Lusty and Deborah Eastwood explain what to look out for.
‘Diseases or complaints of growth’ were first described by Duchamp (1) in his treatise ‘Maladies de la Croissance’ in 1832. These aches and pains were initially thought to be manifestations of rheumatic fever requiring prolonged bed rest, a notion that gave way to the ‘fatigue theory’ proposed by Bennie in 1894 (2). This declared that ‘little legs have performed an unusual amount of work during the day’ and may therefore hurt at the end of the day. In 1939 Hawksley (3) emphasised the occurrence of limb pains in healthy children at night or at the end of the day, and by 1951 Naish and Apley (4) had developed clear diagnostic criteria for growing pains (see Table 1 below) to explain the vague, nebulous discomfort in the lower limbs of which children frequently complain and for which no cause can be found.
Symptoms can range from mild aches at night to severe cramp-like pain waking the child from sleep. The pain is normally felt in the thighs and/or the calves and very rarely involves upper limbs. The pain is bilateral, and is not usually localised to a joint such as the knee. Daytime activities and sports do not obviously exacerbate the pain. Clinical examination and investigations prove normal.
How common these symptoms are depends on the diagnostic criteria used and the population studied. Naish and Apley (4) examined 720 healthy English children and found that, defined by their strict criteria, 4.2% had a history of growing pains. Other researchers came up with rather different results (see Table 2 below). The age group most commonly affected is three to five year olds and symptoms are rare beyond the age of 12. A higher incidence is reported in girls (although this is not statistically significant).
Growing pains are not caused by growth, but they do cease when growing stops. The growth rate slows from three years of age onwards and only picks up again with the onset of puberty, so it is declining at the time when most of these patients present. The growth rate of children who have growing pains is no different from that of those who do not (5). Hawksley, in 1938 (3), suggested that there may be an anatomical basis for these pains and if the children were properly examined for scoliosis, a foot deformity or rotational abnormalities of the lower limbs, an anatomical cause for the pains would be found. However, this theory has fallen out of favour because numerous other authors, on close examination of their patients, have failed to find any such significant anatomical variations.
|History of pain lasting for three months or more|
|Symptom-free intervals of days, weeks or months|
|Pain at the end of the day or waking the child at night|
|Pain of significant severity to interrupt activity such as sleep|
|Pain not specifically related to the joint|
|Normal physical examination|
|Normal laboratory tests and X-rays|
In 1951 Naish and Apley said: ‘Physical growth is not painful but emotional growth can hurt like hell.’ They described a triad of abdominal pains, headaches and limb pains, with psychological profiles showing emotional disturbances in all their patients. More recently, it has been suggested that the muscle pain arises from local ischaemia (shortage of oxygen) in a muscle, causing muscle contraction which leads to capillary constriction, a build-up of irritative metabolites and further local ischaemia and pain, all setting up a vicious circle of spasm and pain. Hashkes (8) looked at 44 patients and 46 controls and found a lower pain threshold in children with growing pains, with increased muscle tenderness similar to fibromyalgia. He suggested that growing pains may be a variant of a non-inflammatory pain syndrome.
The real difficulty with growing pains is not in making the diagnosis but rather in mistaking the diagnosis and labelling a far more serious pathological process as growing pains (see Table 3, below). Growing pains is, in essence, a diagnosis by exclusion, and common sense (rather than exhaustive investigations) will exclude most worrying diagnoses. But Bennie (2) was correct when he wrote: ‘the longer an MD practises the less the diagnosis of growing pain is made as his diagnostic acumen increases’.
|Author||No in sample||Age||Sample type||Incidence|
|Hawksley, 1939 (3)||505||4 to 14||Hospital patients||33.6%|
|Naish and Apley, 1951 (4)||721||4 to 11||Schoolchildren||4.2%|
|Brenning, 1960 (7)||676||6 to 11||Schoolchildren||19.8%|
|Oster and Neilsen, 1972 (5)||2178||6 to 19||Schoolchildren||15.5%|
|Evans and Scutter, 2004 (6)||1445||4 to 6||Schoolchildren||36.9%|
The clinician’s most important diagnostic tools are a full history and a complete examination of the child. The investigations that are most frequently requested include:
The full blood count will pick up some cases of leukaemia (although a normal test does not rule this out) and most cases of infection, while an ESR will be elevated in inflammatory and infectious diseases. An X-ray will only show changes in the bone 10 to 14 days after the start of the pathological process unless there is a fracture.
In MacArthur’s study (9) of why physicians ordered tests in growing pains, one of the main reasons given was ‘family reassurance’. Investigations, however, should be used primarily to help answer a question and lead to a diagnosis. If there is diagnostic uncertainty then, depending upon the individual child’s mode of presentation, various specialist tests may also be requested including additional blood tests, radio-isotope bone scans, computer tomography and MRI scans.
Growing pains are benign, self-limiting conditions usually resolving within two years. Many treatments have been advocated, including simple reassurance, vitamins, salicylates (such as aspirin), friction rubs and psychiatric counselling. The most popular treatment regimes are reassurance, ‘home’ massage programmes and analgesia. Baxter (10) suggested that parents perform a 10-minute, twice-daily programme of stretches to the quadriceps, calf muscles and hamstrings. She demonstrated an impressive improvement in the number of episodes of pain per month and the time to resolution. Whether this was the result of a physiological change in the muscles by stretching promoting improved blood supply, or increased parental-child interaction and psychological support is not clear. However, it is a benign intervention with good reported results that can certainly be recommended.
It is beyond the scope of this article to discuss every possible differential diagnosis for knee pain in the growing child. However, there are a few important conditions that must be recognised.
|Anterior knee pain||
|Pain referred from the hip||
Hilton’s Law states that the nerves that supply the muscles that move a joint also supply sensation to the joint itself. The muscles that move the hip also move the knee, so pain in the knee may be felt in the hip and vice versa. Every orthopaedic trainee has drummed into them the need to examine the hip when a patient complains of a knee problem but we still see too many missed diagnoses at tertiary referral centres (see case study below).
A practitioner rarely misses a child with an infection who is acutely unwell, but children may present with knee pain shortly after having had a respiratory illness treated with antibiotics and much more subtle signs of their bone or joint infection, which can be easily missed.
Pain that is not localised to a joint, but is well localised to one part of the bone such as the distal femur or proximal tibia may indicate an infection or tumour. Soft- tissue changes may be present, although true cellulitis is uncommon. If the pain is localised to a joint and if that joint is irritable, an infection of the joint (septic arthritis) must be considered. A Brodie’s abscess (a small chronic localised bone infection), an osteoid osteoma (benign bone tumour) and some malignant tumours will all give night pain. The pain from osteoid osteoma is very responsive and specifically so to anti-inflammatory medication. Important tests for diagnosis include a full blood count and ESR, as well as X-rays and occasionally a bone scan.
These conditions are readily diagnosed with history and examination and, in the typical patient, require no further tests. There should be no history of trauma, and the knee pain should be well localised. It is often bilateral, although not symmetrical (see fig 1 below).
i) Patellofemoral pain:This used to be called chondromalacia patellae. Patients have anterior knee pain localised to the patellofemoral joint. They may have crepitus from the joint and complain of giving way. Swelling is not a prominent feature. Patients usually respond to a stretching and strengthening programme concentrating on the quadriceps and hamstring muscle groups, perhaps particularly the vastus medialis group.
ii) Osgood-Schlatter disease:In reality this is a condition rather than a disease. Patients present with pain, tenderness and swelling localised to the proximal tibial tuberosity. The pain is exacerbated after activity and resisted knee extension. It is thought to be caused by a repeated traction injury on the developing ossification centre of the tibial tuberosity, which causes inflammation and structural change. However, MRI studies have shown changes within the patellar tendon consistent with tendinitis. Treatment is controversial. The currently recommended treatment is RICE: rest, ice, compression and elevation, with the use of analgesic and non-steroidal anti-inflammatory medication as necessary. The condition should resolve within two years.
An important concept in management is that of ‘relative rest’. We allow our patients to continue playing sport as symptoms permit. We have no objections to the intermittent use of strapping or bracing, particularly if some limitation of flexion may give symptomatic relief during sport. Occasionally a short period of immobilisation will be needed. Persistent cases may have a symptomatic mobile ossicle within the tendon This can be removed surgically but only in exceptional cases as the risks of surgery can outweigh the benefits.
iii) Sinding-Larsen-Johansson disease:This condition is analogous to Osgood- Schlatter disease but affects the inferior pole of the patella. It, too, usually responds to non-steroidal medications and RICE, but stretching and strengthening regimes may also help.
iv) Discoid meniscus:In about 2% of Western populations, the meniscus, rather than having a normal semi-lunar structure, is disc-shaped. Most patients present in childhood; a few as adults, and a few remain symptom-free. Children present between the ages of six and 12 years, complaining of popping, snapping or instability in the knee. If symptoms are troublesome, these patients need to be investigated with X-rays and MRI scans. The treatment involves arthroscopic surgery.
v) Meniscal cysts:Patients present with symptoms and signs typical of meniscal pathology. They often have pain over the joint line and sometimes associated swelling. Classically the swelling from the cyst decreases in size with knee flexion and increases with extension. MRI can confirm the presence of a cyst and identify whether there is a concomitant meniscal tear, as well as excluding other pathologies.
If the patient has a history of significant trauma prior to the onset of their growing pains, they should undergo a thorough examination, possibly in a hospital setting, to rule out fractures, meniscal injuries and disruption to the ACL complex. Ligament injuries are unusual in this age group, as ligaments are stronger than the bones they insert into, under most conditions.
There is evidence that cases of persistent growing pains are being diagnosed in later life as restless leg syndrome (11), an autosomal dominant condition (it runs in families) originally thought mainly to affect the middle aged and elderly. Patients complain of leg discomfort and the urge to move their legs. Symptoms are worse at rest or at night. Patients have periodic involuntary limb movements in their sleep and when awake. These movements, when recorded on polysomnography, are diagnostic. Caffeine and chocolate have been reported to exacerbate the condition; some patients respond to antiepileptic or dopaminergic drugs.
Patrick Lusty is an orthopaedic specialist registrar based at the Royal National Orthopaedic Hospital in London
Deborah Eastwood MB FRCS is a consultant paediatric orthopaedic surgeon, based at the Hospital for Sick Children, Gt Ormond St, London and the Royal National Orthopaedic Hospital. Her special interest is in working with disabled children in sports and non-sporting contexts
Illustrations by Viv Mullett