The exact number of patients with MMC in Norway is not known. For this study, the patients were recruited from TRS, a National Centre for Rare Disorders in Norway [¹⁴]. Registration at the centre is voluntary. Most of the registered patients with MMC had been investigated at the centre during the preceding years and had lumbosacral MMC. By the start of the study in August 2003, a total of 193 patients were registered. All of them were included and received written information about the purpose of the study and the questionnaire. Of these, 87 patients (45%) gave their written consent for participation and completed the questionnaire (Table 1).

The study followed a cross-sectional design with a questionnaire-based systematic telephone interview. This means, that the patients were interviewed only at one point in time without a follow-up interview. Consequently, risk factors for sores can only be investigated through associations and not as causal factors, which would require a longitudinal effect study.

The questionnaire was designed in collaboration with the Head of the National Spina Bifida and Hydrocephalus Users' Organization, both concerning the topics and in the formulation of questions [¹⁵]. The questionnaire was not formally validated. It contained a total of 225 questions on five pages. The questions required mostly "yes/no" answers, although some of them contained several categories. Most of the topics were assumed to be potential risk factors for pressure sores.

The essential topics of the questionnaire are summarized in Table 2. The last part of the questionnaire allowed for open comments about personal experience of sores and individual recommendations concerning their prevention. The questionnaires were sent to the participants two weeks in advance of the telephone interview. The participants were encouraged to collect information concerning the questions from their families and doctors. Both the patient and the interviewer had the questionnaire at hand during the interview. Most of the patients had been at the centre several times and were known by the interviewer, P?l-Erik Plaum. The interviews were performed between August 2003 and December 2004.

The "European Pressure Ulcer Advisory Panel" defines pressure sores as an area of localized damage to the skin and underlying tissue caused by pressure, shear, friction and/or a combination of these [¹⁶]. Pressure sores are classified as follows: grade 1: non-blanchable erythema; grade 2: abrasion or blister; grade 3: superficial ulcer; grade 4: deep ulcer. The questionnaire inquired solely about skin alterations corresponding to grades 3 and 4, because the mild skin alterations were assumed to be difficult for the patients to recognize.

The medical condition and function of the participants were assessed in terms of mobility and sensory function, the presence of stoma, the use of diapers, neuropsychological function and the presence of additional neurological diagnoses. The most apparent neurological impairment of MMC is lower limb paralysis and sensory loss. However, it was not possible to examine the patients neurologically and consequently the exact anatomical level of the lesion could not be obtained. To acquire general information about the mobility a number of questions were asked (Table 2). Concerning the sensory function the patients were asked if they had reduced or missing sensibility from buttocks to knee, from knee to toes, or on the whole leg, both right and left. Because the additional neurological diagnoses (tethered cord and Arnold Chiari malformation) are often asymptomatic and not known to the patients, we asked whether they had surgical treatment for these conditions.

The patients were categorised in three groups: group (1) reporting sores at the time of the interview, group (2) reporting sores in the last five years, but not at the time of the interview, and group (3) reporting of never having had sores. The localization and the duration of the sores in group (1) were recorded.

Associations between potential risk factors and sores, as categorised in these three groups, were tested using contingency tables and the Pearson chi-squared test. A p-value of 0.05 or less was considered statistically significant. No correction for multiple testing was done.

The study was approved by the Regional Committee for Medical Science Ethics and the National Organization of Spina Bifida and Hydrocephalus [¹⁷].

The general characteristics of the 87 participants, of the non-responders and of the total population are presented in Table 1. No significant differences in age and gender distribution were found between the participants and the non-responders. Information concerning the living factors, medical condition, function, and some aspects of sore management is summarized in Table 2. The main findings are that 50% of the patients lived alone, 78% were unemployed, and that 56% were permanent wheelchair users. Sensory deficits, reported in 95% of the study group, were most obvious distally in the legs and symmetrically distributed (Figure 1). Concerning the neuropsychological deficits, 63% of the patients had simultaneous capacity deficit, 62% had concentration difficulties, 60% had learning difficulties, 51% had memory deficit, 51% had orientation problems and 6% had speech deficit. Shunted hydrocephalus was present in 62% of the patients, and 18% had surgically treated Arnold Chiari malformation.

Sores were present at the time of the interview in 26 patients (30%; 95% CI = 20 ? 40%), present in the last 5 years, but not at the time of the interview in 45 patients (52%; 95% CI = 41 ? 62%), and 16 patients reported never having had sores (18%; 95% CI = 11 ? 27%).

In all cases, sores were associated with the degree of sensory deficit: 19 patients reported sores in regions with missing sensibility, 7 patients reported sores in regions with reduced sensibility, and none of the patients reported sores in regions with normal sensibility. The significant associations are shown in Table 2. Pressure sores were significantly associated with memory deficit and with surgically treated Arnold Chiari malformation, p = 0.02 for both associations. Another significant association was found between previous sores and current sores, p = 0.004: all the 26 patients with current sores had also had sores at some point in the previous five years. The other parameters gave no significant association.

All the 87 patients reported that they inspected their skin themselves to see if they had sores. However, 18 patients (21%) also had other people who inspected their skin, 11 were family members and nine were health professionals (two of the patients had skin inspection by both, family members and professionals).

Out of the 26 patients with a total of 41 current sores, 16 had one sore, seven had two sores, one had three sores, and two had four sores (Figure 2). There were 12 sores located on toes, 10 on feet and eight on buttocks. Other localizations were rare. The duration of the sores varied from less than one week to longer than 26 weeks and most of the sores lasted longer than one month with many longer than six months (Figure 3).

The sores were treated in 13 cases by the patients themselves, in 12 cases by professionals and in 5 cases by family members. Consequently, a few cases were treated by several persons. Sore treatment was done daily for 20 patients, three were treated every second day and three had weekly treatment. Out of the 71 patients having sores both currently and previously, 23 had outpatient follow-up because of previous sores and seven due to current sores, 26 (37%) had been examined by a plastic surgeon and 13 (17%) patients with previous sores and two with current sores had undergone surgery. Many of the patients had received follow-up over many years. Patients were encouraged to note their personal experiences for treatment and avoidance (Table 3). It was noticed that the sores in some patients were localized several times at the same place. Several patients reported less sores while using penicillin for treatment of a urinary tract infection.

For age and gender, no significant differences were found between the participants and the non-responders. In the registry used for this study, there was no systematic data on neurological function, the level of the MMC or the presence of additional neurological diagnoses. However, in a previous study, the prevalence of the additional neurological diagnoses of MMC in Norway was 64% for shunted hydrocephalus, 13% for syringomyelia, 80% for tethered cord and 43% for Arnold Chiari malformation [¹⁸]. This indicates that the study population used here was representative for the patients with MMC in Norway concerning the severity of the condition. From the reported distribution of sensory and motor deficits, it seems likely that all the patients had lower thoracic or lumbosacral myelomeningocele.

The prevalence of sores at the time of the interview found in this study corresponds with previous findings in Norway from 1996 [²], where the sore prevalence was estimated to be 35% and to a Spanish multicenter study from 1993, where the sore prevalence was 31.7% [⁵]. Variations in sore prevalence in different studies may be explained by differences in population age, different inclusion criteria, and different target groups by the medical care systems [^3-8].

All the data were acquired through a questionnaire-based interview. To help with reliability, the questionnaire was sent prior to the interview and the patients were encouraged to gather information before the interview was performed. A weakness of the study is that self-reporting was used to describe clinical parameters, and that the data were not validated. Therefore statements on associations between clinical factors and sores should be interpreted cautiously. The p-values of the associations found between sores and memory deficit and surgically treated Arnold Chiari malformation were not very strong and they have not been adjusted for multiple tests. If a physical examination had been used, the data would have been more reliable.

The aim of this study was to identify risk factors for sores. Sores were invariably associated with sensory deficit. This finding is not unexpected because reduced superficial sensation increases the risk of mechanical and thermal skin damage. Impairment of the autonomic innervation alters blood circulation, temperature homeostasis and healing processes. This could explain why sores were mostly localized on feet and toes. This corresponds with a previous study, which showed sores on the plantar surface of the metatarsal heads and of the heel in 75 and 64%, respectively, and sores were related to the absence of plantar sensation [¹¹].

Mobility function was not associated with sores in contrast to other studies, where the motor level was related to sore incidence [⁷,⁹]. This may be because the questions on mobility function were insufficiently precise.

Among the cognition deficits recorded, sores were associated with memory deficits. This seems evident because intact memory function is essential for skin inspection routines, sore treatment, and prophylactic actions such as regular "sit-ups". Only five patients reported problems with speech. From the literature, it is well known that individuals, mostly children, with MMC have a certain cognitive profile, characterised by reduced memory, cognitive processing, arithmetic, visual perception/construction and fine motor coordination [^19-24], but also with approximately normal verbal ability [¹⁹,²⁰,²³,²⁵].

Of the additional neurological diagnoses often occuring with MMC, sores were significantly associated only with surgically treated Arnold Chiari malformation. On the other hand, somewhat unexpectedly, hydrocephalus, either shunted or not, gave no association. This can be explained by recent research showing that patients with MMC and Arnold Chiari malformation seem to have a distinct cognitive profile characterised by impaired visual analysis and synthesis, verbal memory, and verbal fluency [²⁶,²⁷].

Surprisingly, our study found no associations between sores and living factors such as consciousness of nutrition, smoking, living together with other persons, employment, regular physical activity and physiotherapy. Nor were gender, age and body mass index associated with sores.

All the patients with current sores had also had sores at some point in the last five years. This indicates that patients with current sores often have a history with repeated sores. The high prevalence rate over several years could indicate insufficient diagnosis and treatment. In the literature, recurrence rates greater than 80% are reported which led to a new approach with an education program in personal skin and self-care after reconstructive sore surgery that could reduce sore recurrent rate to 25% [²⁸]. In this context, the comments about personal experiences revealed that sores typically were on the same area and emerged gradually with the risk of becoming chronic and more severe. The patients' recommendations for the sore management show that many of them have been informed about useful strategies. To what degree these activities were performed and were effective is unknown. Possibly improved skin inspection, especially by other persons, might contribute to a reduced prevalence.

We thank Eli Skattebu, the Head of the National Spina bifida and Hydrocephalus Users Organization, and Svend Rand-Hendriksen from TRS for collaboration in elaborating the questionnaire, the National Health Organization for providing information about spina bifida incidence and prevalence in Norway, and Kristin Brautaset from TRS for language assistance.

When the sum of the "Yes-" and "No-column" was less than 87 (100%), the parameter was unknown for the participant. The last column presents p-values for associations between the parameters and sores. Remarks: n.s. = not significant; # = p-values were not calculated due to low numbers.