Prolonged, widespread, disabling musculoskeletal pain of adolescents among referrals to the Pediatric Rheumatology Outpatient Clinic from the Päijät-Häme Hospital District in southern Finland

1 Introduction

The epidemiology of prolonged non-inflammatory widespread pain varies by age, determination and country among adolescent populations [1], [2], [3], [4], [5], [6], [7]. In southern Finland the prevalence of widespread pain varies from 3% to 7.5% [7], [8], whereas in Norway, musculoskeletal pain in three or more locations was reported by 8.5% of adolescents aged 13–18 years [4]. Although among 18-year-olds in European countries and Israel, young people report that chronic pain affect their social and working lives, very few were managed by pain specialists [1].

All in all, the prevalence of musculoskeletal pain among adolescents varies according to different studies from 4% to 40%, multiple pains ranging from 4% to 49%. The pain prevalence rates for patients were higher among girls and usually increased with age [6].

Pediatric rheumatologists review a large number of patients with a wide variety of causes of musculoskeletal pain [9]. In USA, musculoskeletal pain was the most common reason for 226 out of 414 children (55%) referred to pediatric rheumatologist for over 3 years [10].

The economic impact of prolonged pain is considerably high. The financial burden of adolescent chronic pain on the UK economy for 1 year, 2005, was calculated to be £3,840 million. The mean cost is approximately £8,000 per year per adolescent and their family, including direct and indirect costs, the variation being large from £200 to over £40,000 per year. The costs were higher for patients with non-inflammatory pain than for those with inflammatory pain [11].

It is important to identify those patients who need treatment. A nine-item screening tool, the Pediatric Pain Screening Tool (PPST), identifying prognostic factors associated with disabling prolonged pain, was introduced by Simons et al. [12]. This screening tool includes nine items divided into two subscales: physical and psychosocial subscale. These items measure functional disability, pain catastrophizing, fear of pain, anxiety, and depressive symptoms. Patients were classified into low-, medium-, or high-risk groups. According to Simons et al. patients with total score 0–2 are low risk patients and patients with a total score of ≥3 are at medium- or high-risk. Of these, patients that scored 0–2 in the psychosocial subscale were medium-risk-patients and ≥3 in the psychosocial subscale were classified as high-risk patients.

At the 4-month follow-up, the PPST was valid and facilitated risk stratification to identify patients who need effective pain treatment [12].

Chronic idiopathic musculoskeletal adolescent pain, especially in three or more sites, has been reported to have a major impact on several areas of daily living [4]. Hoftun et al. defined the subjective disability index by: difficulties in sitting during a school lesson, attending daily activities during leisure time, sleeping difficulties, problems walking more than 1 km, and problems during a physical exercise class. These were assessed to count the subjective disability index. This index was higher in girls and increased with the frequency of pain and the number of pain locations [4].

The aim of the present study was to determine the incidence of prolonged, disabling, non-inflammatory, musculoskeletal pain from the referrals from the Päijät-Häme Hospital District to the Pediatric Rheumatology Outpatient Clinic of the Päijät-Häme Central Hospital, Lahti, and to describe the patient material. Patients with generalized, prolonged, musculoskeletal pain who had daily problems regarding various areas of life were the target of our search for rehabilitation of patients.

Additionally, our purpose was to find factors that indicate the need of targeted and specialized pain treatment. We aimed to test whether our patients fit the PPST risk group stratification by Simons et al. and whether our patients’ functional disability correlates with Simon’s risk stratification.

2 Patients and methods

This study took place in the Päijät-Häme Hospital District in southern Finland. The district consists of both rural and urban areas, with approximately 220,000 inhabitants. All patients were recruited from referrals to the Pediatric Rheumatology Outpatient Clinic at Päijät-Häme Central Hospital.

From 2010 to 2015, we received 508 patients to our pediatric rheumatology clinic. Out of all referrals between 2010 and 2015, patients who were diagnosed with juvenile idiopathic arthritis (JIA), connective systemic rheumatic diseases, transient synovitis of the hip, growing pains or another disease explaining the prolonged pain were excluded from the study, thus leaving the 508 patients. The patients, among the 508 referrals, were aged from 1 year to 17 years.

First, the patients and their accompanying parent/parents were interviewed by a nurse specialized in pediatric rheumatology.

The patients were then examined by a pediatric rheumatologist together with a specialized physical therapist. In addition, all patients had a personal physical therapist’s consultation.

Chronic illnesses were excluded by means of laboratory tests, with the following assays analyzed from all patients: erythrocyte sedimentation rate, C-reactive protein, complete blood count, creatinine, alanine aminotransferase (ALT), antinuclear antibodies, rheumatoid factor, citrullinated peptide antibodies, anti-tissue transglutaminase IgA antibodies, IgA to exclude its deficiency, HLA-B27, Saccharomyces cerevisiae antibodies (ASCA) to exclude Crohn’s disease, and borrelia burgdorferi antibodies.

Between September 2010 and October 2015, 34 out of the 508 patients were selected by a multiprofessional team to participate in a prolonged pain rehabilitation program in Lahti. The team included two pediatric rheumatologists, a clinical psychologist, a physical therapist, and a nurse. The inclusion criteria for the intensive outpatient pain program for adolescent prolonged multisite musculoskeletal pain were: prolonged musculoskeletal pain lasting over 3 months, school absence, sleep or mood disturbances, problems with physical exercise, or difficulties with daily activities during leisure time. All other 474 patients had minor problems of pain and did not have daily problems because of pain.

One patient was excluded because he was only 10 years old and was too young for rehabilitation. His treatment continued normally at the outpatient pediatric rheumatology clinic. One patient, a boy of 16 years with CRPS, refused to attend the rehabilitation thus the study group being 32 patients.

In this group of 32 patients, with pediatric rheumatologist’s examination duration of pain, laboratory tests and by contrast enhancement magnetic resonance imaging (MRI), other chronic illnesses were excluded. Twenty-seven out of 32 patients (84%) were examined by MRI to exclude chronic arthritis or other illnesses and to confirm idiopathic musculoskeletal pain with no other cause. All MRI examinations of one most painful joint were normal. We did not use whole body MRI scan.

Our target was to indentify patients with prolonged, non-inflammatory, musculoskeletal pain with disabling various problems regarding daily life.

The patient and accompanying parent received both written and oral information about the study. The participants gave their written informed consent.

This study was approved by the Ethical Commitee of Päijät-Häme Central Hospital.

3 Results

3.1 Incidence

Out of the overall 508 referrals, 32 consecutive patients were included to participate in an intensive outpatient pain program for adolescent prolonged, multisite musculoskeletal pain. The inclusion criteria were prolonged musculoskeletal pain lasting over 3 months, school absence, sleep or mood disturbances, problems with physical exercise, or difficulties with daily activities during leisure time.

One of our patients had Marfan syndrome, and four had Ehler-Danlos hypermobility-type-syndrome, and all of these diagnoses were confirmed by a physician specialist in genetics.

During years 2010–2015 the incidence of prolonged, multisite musculoskeletal pain was 42/100,000 patient years (pyrs) (age 13–18; 95% CI 29–60), approximately seven patients in a year. The median duration of pain was 24 months (Interquartile range 38.5).

The incidence of boys was 16 (6–35) and girls 69 (45–101), age at baseline 14.4 (13–17).

According to Simons et al. 19 patients were classified as high-risk (total PPST score ≥3; psychosocial subscale score ≥3), and 13 were classified as medium-risk (total PPST score ≥3; psychosocial subscale score 0–2). None were in the low-risk group (Table 1).

Table 1:

According to Simons et al. we divided our 32 patients into two groups: the medium-risk group (n=13) fulfilling the Simons criteria 3–6 and the high-risk group (n=19) fulfilling Simons criteria 7–9.

	All	PPST		p-Valuea
	n=32	3–6 n=13	7–9 n=19
Girls, n(%)	26 (81)	10 (77)	16 (84)	0.67
Age, mean (SD)	14.4 (1.3)	14.2 (1.1)	14.6 (1.3)	0.30
BMI, mean (SD)	23.3 (5.9)	22.3 (6.6)	24.1 (5.4)	0.41
Duration of pain, mo., median (IQR)	24 (15, 72)	24 (18, 36)	24 (15, 72)	0.32
CHAQ, mean (SD)	0.25 (0.33)	0.09 (0.21)	0.37 (0.36)	0.017
Pain, VAS, mean (SD)	42 (25)	31 (23)	50 (23)	0.031
General, VAS, mean (SD)	28 (19)	20 (18)	33 (19)	0.047
Structured pain questionnaire (SPQ) score	12.5 (6.1)	7.5 (4.0)	15.9 (4.9)	<0.001
Musculoskeletal pain1, n(%)
Neck or shoulder	25 (78)	8 (62)	17 (89)	0.091
Upper ext.	20 (62)	5 (38)	15 (79)	0.030
Chest	10 (31)	2 (15)	8 (42)	0.14
Upper back	18 (56)	4 (31)	14 (74)	0.029
Lower back	22 (69)	7 (54)	15 (79)	0.24
Lower ext.	28 (87)	9 (69)	19 (100)	0.020
Three or more locations	25 (78)	8 (62)	17 (89)	0.091
Headache/migraine, n(%)	22 (69)	7 (54)	15 (79)	0.24
Abdominal pain, n(%)	10 (31)	2 (15)	8 (42)	0.14
Walking test, m/6 min, mean (SD)	577 (58)	575 (58)	578 (59)	0.93
CDI, mean (SD)	10.2 (8.5)	6.9 (7.3)	12.5 (8.8)	0.039
CDI 13-, n(%)	9 (28)	1 (8)	8 (42)	0.033
Smoking	7 (22)	3 (23)	4 (21)	0.89

The differences between the medium-risk and high-risk group, were found in Child Health Assessment Questionnaire (CHAQ), pain visual analogue scale (pain VAS), general visual analogue scale (VAS), and CDI, meaning that the high-risk group had more functional disability, worse general wellbeing, higher pain and depressive scores. The Simons criteria were strongly associated with the frequency of pain, multisite pain (SPQS), and functional disability (CHAQ). Seven patients (22%) were smokers, and 26 (81%) were girls (Table 1).

Two out of the 32 patients did not attend school at all because of pain, 17 had school absences, and 13 attended school regularly.

All patients had pain in more than one body part. Twenty-five (78%) patients had difficulties sitting still during a school lesson, and 16 (50%) had sleep disturbances.

Twenty-five (78%) patients’ pain was increased if they were physically active, 24 (75%) patients worried about pain a lot, 21 (66%) thought that their pain was horrible, 29 (90%) did not enjoy life, and all patients said that pain was a problem for them (Fig. 1).

Fig. 1:

Patient classification according to Simons et al. [12]

Figure 2 shows that the Simons criteria and the structured pain questionnaire score describe functional disability well according to CHAQ (correlation 0.72 (0.50–0.86).

Fig. 2:

PPST and SPQ functional ability according to CHAQ. Correlation 0.72 (0.50–0.86). Blank circles CHAQ 0, black circles CHAQ>0.

Table 2: A patient according only to psychosocial criteria was in the high-risk group if more than three psychosocial criteria were met. Of our patients, 28 (88%) were in the high-risk group, three patients were in the medium-risk group, and none were in the low-risk group.

Table 2:

Psychosocial criteria among our patients according to PPST.

Psychosocial criteria	Frequency	Percent	Cumulative
1	1	3.13	3.13
2	3	9.38	12.50
3	5	15.63	28.13
4	6	18.76	46.88
5	17	53.13	100.00
Total	32	100.00

1. High risk patients: 28 88% (95% CI: 71–96).

One patient fulfilled only one psychosocial criteria, while two patients met three, three patients fulfilled five, four patients met six, and 17 patients fulfilled all five psychosocial criteria.

There were only four patients whose psychosocial score was only 0–2 (Table 2).

4 Discussion

The incidence of prolonged pain among patients referred to a pediatric rheumatology clinic was 42/100,000 pyrs. As a comparison in USA, pediatric rheumatologists reviewed 414 children and 111 (27%) had musculoskeletal pain as an isolated complaint. Only one of them (0.2%) had chronic inflammatory disease [10]. In Canada the proportion of pain syndromes was 15% among diagnoses made in pediatric rheumatology clinic [22]. On the other hand, our patients had severe pain associated disability, meaning that our incidence, which was low compared to those figures from USA and CANADA, is not wholly comparable with studies mentioned above. We did not use the diagnosis fibromyalgia, although it is probable that most of our patients would have fulfilled the criteria for fibromyalgia.

Most of our patients, 26 (81%), were girls as usually in previous studies [6]. Our patients suffered from prolonged pain, and the median duration of symptoms ranged from 15 to 72 months, with a median of 24 months. In the studies by both Sleed et al. [11] and Logan et al. [23], the duration of pain was also approximately 2 years [11], [23]. Our patients had problems with school attendance (78%), which, according to Logan et al. [23], may have a longstanding effect on their lives, and this issue needs to be addressed.

In the current study we did not measure parental factors, which is a limitation in the study. However, according to Hoftun et al. [24] parental chronic pain has been found to be associated with nonspecific pain and especially with multisite pain, in adolescents and young adults, which implies that pain is also a family matter [24]. The burden on the lives of families in attempting to cope with a child with prolonged pain is significant. Adolescents suffering from prolonged pain have various symptoms – sleeping disturbances, problems with school attendance, mood problems/depression, and difficulties in their social life. Pain has a negative impact on the physical and psychosocial well-being of the patients and their families [25].

In a Finnish study, chronic pain showed a strong association with disability retirement [26]. In order to fully understand the pain syndromes in childhood, as well as the risk factors that predict pain in adulthood, we need more longitudinal studies. It is vitally important to target the treatment and rehabilitation as early as possible with preventive measures [27].

The economic burden due to prolonged pain among adolescents is a problem for the patients, their families, and for the economy as a whole. In the US, pain is a considerable problem, with an economically estimated 1.7 million children costing $19.5 billion dollars annually [12]. The costs on the UK economy for 1 year, 2005, were calculated to be £3,840 million. An economic burden of that magnitude is alarming. If these patients are not provided with a treatment that is well-targeted and timed correctly, their personal suffering and economic impact is going to escalate in adulthood [11].

Among our patients, 25 (78%) had pain in three or more locations. Pain-VAS was used as a measure of pain intensity. In the whole sample the pain intensity was 42, but as the patients were stratified into the medium (VAS 31) – and high risk (VAS 50) – groups by PPST there were a significant difference between the groups with the high-risk group having a higher level of pain intensity.

SPQ describes pain frequency and pain location combined. Ståhl et al. studied neck pain and found that the frequency of pain reflects well the intensity of pain [28]. Moreover, multisite pain and the frequency of pain have been shown to associate with a higher disability index and predict pain persistence [4]. In the current study, PPST and SPQ showed positive correlation with functional disability reported by CHAQ.

Simons et al. published a nine-item screening tool to identify factors associated with poor outcome in adolescents with prolonged pain. The screening tool helps to identify at an early stage those patients who need prompt rehabilitation. That tool was proven to be valid in our patient group. It is important to treat patients early because brain imaging shows that brain alterations occur in patients with chronic musculoskeletal pain [29]. In child patients with CRPS these brain changes may be reversible when treated adequately. This emphasizes the possibility of recovery [30].

In a Finnish School Health Promotion Study (SHP) 2015 about 5% of adolescent aged 14–17 attending comprehensive and secondary school smoked. However, the proportion of smokers was higher (10–27%) among those who attended vocational school [31]. In our study, all smokers attended vocational school, and the proportion (22%) was rather similar. Smoking is a risk factor for musculoskeletal pain persistence among girls aged 16 and 18 [32].

The number of our patients was rather small, with only 32 included in the study, and hence the results are preliminary, and must be interpreted cautiously.

Conclusions: The incidence of disabling pain patients in our study was low. Screening tools proved to be valid in our patient material.

Implications: Patients with prolonged musculoskeletal pain are often treated in pediatric rheumatology outpatient clinic. They have pain associated disability covering various areas of life. Methods which provides us sensitive methods to recognize patients with a risk for pain chronification are essential. That should be done already earlier than in our patients, median duration of symptoms 24 months. SPQ score, Hoftun disability index and PPST stratification have been proved to be good tools to identify right patients. Patients need prompt, well-guided and effective treatment, and recognizing the right patients early on is vitally important for a better outcome.