1. Introduction
The International Paralympic Committee, the global governing body of the movement of the athletes with an impairment, organizes, every four years and “parallel” to the Olympic Games, some of the largest sportive events in the world named Summer and Winter Paralympic Games (PGs) [
1]. In the last Summer PGs (Paris 2024), about 4400 athletes participated competing in 22 sports: Para archery, Para athletics, Blind 5-a-side soccer (for athletes with a visual impairment), Boccia, Para canoe, Para cycling, Para equestrian, Goalball, Para judo, Para powerlifting, Para rowing, Shooting Para sport, Sitting volleyball, Para fencing, Para swimming, Para table tennis, Para triathlon, Wheelchair basketball, Wheelchair rugby, Wheelchair tennis, Para badminton, and Para taekwondo. The latter substituted Parasailing and 7-a-side soccer (for athletes with cerebral palsy), which were present at the Paralympic level up to Rio 2016. In the Winter PGs, athletes compete in six sports: Para alpine skiing, Para biathlon, Para cross-country skiing, Para ice hockey, Para snowboard, and Wheelchair curling [
1].
To be eligible to compete in Para sports, an athlete must have a diagnosis that leads to a permanent impairment [
2]. This impairment must fall within one of the three main eligible impairment types: vision, intellectual, and physical/motor impairments [
3]. The latter type includes eight impairments: impaired muscle power, impaired passive range of movement, limb deficiency, leg length difference, short stature, hypertonia, ataxia, and athetosis [
2,
3]. Physical impairments are divided into two subgroups: neurological and musculoskeletal impairments [
3]. Among the Paralympic athletes (PAs), the most common health conditions that lead to neurological and musculoskeletal physical impairments are spinal cord injury (SCI) and amputation (AMP), respectively [
3].
Apart from the cardiovascular-related physiopathology determined by the health conditions, several differences exist between PAs with SCI (PAs-SCI) and those with AMP (PAs-AMP). It is known for example that from a cardiovascular point of view, short- and long-term adaptations to sport/exercise occur in PAs-SCI [
4], leading often to a reduced oxygen uptake peak (VO
2peak), the measurement of cardiorespiratory fitness, which is inversely related to cardiovascular risk [
5] and is higher in PAs-AMP compared to those with SCI [
5]. During the Paralympic Games, PAs with neurological health conditions, such as SCI, lost more days per year due to infections and gastrointestinal problems compared with PAs with other health conditions [
6].
On the other hand, regardless of the health condition, it has been suggested that pain occurs in wheelchair users because of incorrect posture, chronic overuse, and obesity [
7]. The latter is a common condition in individuals with SCI [
8] and AMP [
9], and it is one of the most prevalent cardiometabolic risk factors [
5,
8,
9]. Phase angle (PhA), measured by bioelectrical impedance analysis (BIA) and reflecting the extra- and intracellular water content/balance, has been suggested as a discriminator of sarcopenia in chronic musculoskeletal pain patients [
10] and as a marker of oxidative stress [
11].
Compared to healthy individuals, those with SCI have lower levels of exogenous antioxidants, but a similar antioxidant response to exercise [
12]. In a previous study, we observed that post-exercise ketosis was associated with a reduced antioxidant response after a simulated wheelchair basketball match, with no differences in Wheelchair Basketball Athletes (WBAs) with different health conditions [
13].
Although concerns about carbohydrate (CHO) intake and glycogen stores have been raised in athletes with SCI [
14], limb deficiency determining muscle asymmetry should affect glycogen stores. Muscle asymmetry is related to pain in individuals with AMP [
15]. It has been suggested that the Mediterranean diet (Med-D) [
16,
17] and plant-based diets [
18,
19] reduce musculoskeletal pain. Moreover, to reduce cardiometabolic risk, the Med-D has been suggested for individuals with SCI [
20] and with lower limb amputation (LLA) [
21]. A prospective intervention (calorie-restricted Med-D and circuit resistance training) in a cohort study involving 20 individuals with SCI reduced body mass index (BMI) and total fat mass, and improved glucose regulation, insulin sensitivity, and lipid profiles [
22]. Additionally, resting energy expenditure (REE), fat oxidation, cardiorespiratory fitness, and dynamic strength increased [
22].
Despite the potential of the Med-D and plant-based diets to reduce both the cardiovascular risk and musculoskeletal pain, the influence of the health conditions on digestive functions is among the determinants of the dietary behaviors of wheelchair users with both SCI and LLA [
23].
It has been stated that energy balance, a fundamental health related parameter in PAs, is extremely difficult to be assessed [
24] due to the measurements of both energy expenditure for the wide ranges of types, intensities, and durations of practiced sports [
25,
26] and activities of daily living [
27] and energy intake for the physio-pathological characteristics of some PAs, such as those with SCI [
28]. Moreover, high variability in resting metabolic rate exists in PAs, as well as dietary restrictions related to the health condition [
28]. In particular, individuals with restrictions on eating due to gastrointestinal symptoms require an evaluation for orthorexia [
29].
It is known that dietary advice for individuals and PAs-SCI requires particular attention [
30,
31,
32], whereas an unresolved question is whether PAs-AMP should follow nutritional advice as able-bodied athletes. Indeed, at our best knowledge only two studies have specifically evaluated PAs-AMP [
33,
34]. In these papers, the PAs of both the Brazilian Wheelchair Women’s Basketball Team [
33] and the male Brazilian Amputee Soccer Team [
34] had lower CHO intake compared to the CHO recommendations for able-bodied athletes. On the other hand, Scaramella et al. [
35] did not give conclusive recommendations for PAs-AMP. Indeed, from one point of view, they [
35] suggested that appropriate CHO intake should be based on the lower end limit of the CHO range of able-bodied athletes for PAs with less active muscle mass (i.e., those with SCI or with double leg amputees) [
35]. However, from another point of view, PAs-AMP should increase their CHO needs, because the possible inefficiency of movement of ambulant athletes with AMP may increase glycogen utilization [
35].
This study aimed to deepen the knowledge on the appropriate nutritional regimen of PAs-AMP, comparing PAs-SCI with PAs-AMP using a multidimensional approach. To accomplish this aim, a descriptive cross-sectional observational study was conducted [
36,
37] with a mixed quantitative methodology (questionnaires, fitness and clinical measurements) [
36] on veteran PAs of the “Gruppo Sportivo Paralimpico della Difesa” (GSPD) [
38]. We hypothesize that, despite the expected differences in cardiorespiratory fitness, gastrointestinal symptoms, and eating behaviors, both PAs-SCI and PAs-AMP can have similar dietary habits and post-exercise glucose and ketone responses. Therefore PAs-AMP would necessitate a tailored nutrition plan [
36,
37] based on recommendations similar to those for PAs-SCI [
14,
30,
31,
32,
35] rather than those for able-bodied athletes.
4. Discussion
The multidimensional evaluation through questionnaires and clinical and functional measurements carried out in the present study in veteran athletes with physical (motor) impairments determined by different health conditions (SCI and AMP) showed a wide range of responses and a great overlap of results. Although the PAs in the present study presented the expected differences in cardiorespiratory fitness (lower in PAs-SCI than PAs-AMP) and gastrointestinal symptoms (NBD scores higher in PAs-SCI than PAs-AMP), adherence to the Med-D, lifestyle characteristics (sedentary behavior and sleeping time), and the number of painful sites were not different. Moreover, similar values were found for the nutritional status and dietary intake, basal metabolism, and glucose and ketone levels after CPET, and body composition in the parts of the body without impairment. Therefore, we verified the hypothesis that athletes with an impairment need a tailored nutrition plan based on a comprehensive clinical and functional assessment that also includes and evaluation of eating behavior (e.g., eating habits, preferences, and orthorexia).
The Paralyzed Veterans of America (PVA) dietary criteria include a nutrition plan similar to the Med-D (high levels of whole grains, fruits, vegetables, legumes, and low-fat dairy products, and low levels of red meat and sugar) [
56]. In the present study, adherence to the Med-D was equal to about 50%, comparable to that observed in international-level WBAs [
39] but with lower values of dietary ORAC, similar to those reported in obese individuals with low adherence to the Med-D [
57] and definitively lower than those found in professional cyclists during training [
58]. A significant negative correlation between the Med-D adherence and the ORTO-15 score has been previously reported in both professional (r = −0.365) and recreational (r = −0.309) able-bodied athletes [
59]. Accordingly, the already quoted previous study of ours revealed through Spearman’s correlation data on WBAs indicated that WBAs with a low ORTO-15 score (high orthorexia) had a high MEDScore (r = −0.539) [
39]. In the present study, we confirmed that Spearman’s correlations showed that PAs with high orthorexia, assessed through ORTO-15, had high adherence to the Med-D. ORTO-15 was also related to NBD scores and ORAC. It has been suggested that ORTO-15 may be related to a commitment to wellness in WBAs [
39]. Contrary to ORTO-15, which can reveal positive healthy characteristics, in university students, the ORTO-7 was more specific than ORTO-15 in assessing orthorexia nervosa, being independent from confounding variables such as body image concerns, distress, appearance, fitness, and health orientation [
53]. In our study, both ORTO-15 and ORTO-7 inversely correlated with EnI (ORTO-15 r = −0.542, ORTO-7 r = −0.624) and PhA of the dominant arm (ORTO-15 r = −0.579; ORTO-7 r = −0.500). Using ORTO-7, the negative correlation between ORTO-15 score and Med-D adherence was not found. Accordingly, Med-D adherence was not related to ORTO-7 in gym attendees with an FM% below 17 [
39]. In the present study, the differences between groups in FM% did not reach statistical significance, but the FM% of PAs-SCI was found to be higher than previously reported data for PAs-SCI [
60]. On the other hand, the FM% of PAs-AMP was comparable to previously reported data found in PAs with above-knee amputation (AKA) or below-knee amputation (BKA) [
61]. We must stress the fact that not only the PAs-SCI but also the PAs-AMP displayed FM% higher than alpine and Nordic skiers and Para ice hockey players [
45]. Indeed, the PAs in the present study, regardless of the health conditions, had an FM% similar to the Para Curlers who had an FM% equal to 26.2 ± 7.74% [
45].
A recent position statement based on expert consensus suggests that the health and fitness evaluation of the preparticipation screening should include nutritional and body composition assessments and a daily energy expenditure (DEE) evaluation [
24]. The energy expenditure of individuals with a locomotor impairment in both physical exercise and Paralympic sports vary widely depending on the type of physical activity, the actual intensity, and the duration of both exercise and sports [
25,
26]. Mean energy expenditure, for example, in sitting intermittent sports (aerobic and anaerobic mixed metabolism), such as wheelchair fencing, wheelchair tennis and wheelchair basketball, and endurance sports, such as sitting Nordic skiing and long-distance wheelchair racing, ranges between 7.21 and 11.7 metabolic equivalents of tasks (Mets), respectively [
25]. The assessment of DEE becomes even more difficult during sport training in mixed metabolism sports, such as wheelchair basketball (circumstance sports) and during technical training. A DEE assessment in alpine sports training can also vary widely. Furthermore, the assessment of DEE is also dependent on the activities of daily living, which are difficult to assess [
27]. Due to the difficulty in the evaluation of energy balance [
24], a strict periodic evaluation should be carried out, addressing the point of tailored nutrition advice.
In the present study, the fat %EnI was higher than the reported one in high-performance PAs [
62], whereas the observed low intake of CHO is consistent with the previous literature related to other PAs [
13,
63,
64]. In a review on CHO consideration for PAs, it has been pointed out that the CHO oxidation rate was lower in PAs-SCI compared to able-bodied athletes, and that they have a greater dependency on CHO timing, due to the reduced glycogen storage capacity [
14]. Accordingly, we proposed that elite WBAs with relatively low CHO intake could be at risk of both malnutrition and post-exercise ketosis [
13]. Furthermore, it has been suggested that able-bodied individuals, following a session of aerobic exercise, presented increased TAC and uric acid levels as a protective reaction against oxidative stress [
65]. In the present study, after CPET, both salivary TAC and blood uric acid levels increased, with no differences between PAs-AMP and PAs-SCI in the antioxidant response to CPET, as well as in glucose and ketone levels after CPET.
Positive correlations between TAC and PhA have been observed in patients with chronic kidney disease and older women [
11]. In individuals with chronic musculoskeletal pain, the proposed cut-off value for PhA discriminating sarcopenia was 5.1 for men [
10]. In the present study, the mean PhA value was below the cut-off for sarcopenia [
10] only in the lower limbs of PAs-SCI.
On the other hand, the mean intake of proteins normalized for BM (mean values ranging from 1.2 to 1.4 g/kg BM) was adequate for PAs (1.2–1.7 g/kg BM) [
20]. Although a high-protein intake diet might have a negative effect on kidney function in individuals with SCI, malnutrition can increase the risk of pressure ulcers [
32]. The latter increases protein needs in sedentary individuals with SCI (at least 1.25 g/kg BM) for wound healing [
32]. Furthermore, because stump ulcers are common problems in amputees using prosthetic limbs [
66], a diet rich in protein would be advisable.
In veterans with unilateral BKA (97.9% with a prosthetic limb), a high prevalence of pain was reported (stump pain 84.2%, low back pain 78.1%, and knee pain 54.7%) and low back pain was higher in amputees with stump pain [
67]. Comparing reported pain, we observed a high prevalence of both lumbar pain and use of anti-inflammatory drugs in PAs-AMP compared to PAs-SCI, whereas the use of analgesic drugs was comparable. Chronic pain is common (80%) in individuals with SCI [
68] and it is known that neuropathic pain in both lower trunk and legs and spasticity can be observed in individuals with SCI [
69]. In a previous study, the prevalence of neuropathic pain was 51% and 38% of patients with cervical and thoracic SCI, respectively, and 85% of patients with thoracic SCI and 87% of those with cervical SCI experienced spasticity [
69]. Overall, in our study the prevalence of musculoskeletal pain at any location was higher (96%) than that previously reported in wheelchair users (50%) [
7]. In both groups, we found the well-documented high prevalence of shoulder pain in wheelchair athletes (68%) and nonathletic wheelchair users (67%) [
70]. Although it has been suggested that shoulder pain has a significant impact on the range of motion, leading to functional limitations [
7], the absence of relationship between number of regions with reported pain and VO
2peak suggested that pain did not affect the execution of CPET on the arm cranking ergometer.
Health conditions, the type of impairment, and level of lesion impact VO
2peak, and therefore the central (heart) and peripheral (vascular and muscular systems) short- [
4] and long-term [
71] adaptations to exercise differ when comparing PAs-AMP and PAs-SCI [
4,
71]. For this reason, VO
2peak, which has an impact on performance in endurance, power, and intermittent (aerobic-anaerobic alternated) sports [
45], shows also an inverse relationship with cardiovascular risk factors [
5]. VO
2peak is therefore a fundamental component of physical fitness to be assessed in PAs competing in different sports [
72]. In our study, cardiorespiratory fitness (VO
2peak and peak power) was higher in PAs-AMP compared to PAs-SCI. PAs-AMP had VO
2peak values consistent with the values of the Italian National WBAs [
13,
26]. Because the intensity and energy expenditure of a sport discipline has a strict relationship with VO
2peak [
26], PAs competing in different sports display different VO
2peak values [
26,
45,
72]. Therefore, the major limitation of our study is that the limited sample size prevented us from pooling athletes in sport discipline groups.
The first point of strength of the present paper is that we collected parameters with non-invasive (VO
2, HR, and saliva) or minimally invasive (capillary blood) methods, making the study easily transferable on a large-scale during training camps. The second point of strength is that performing the test during training camps allowed us to measure BEE after waking up and before getting out of bed by indirect calorimetry, according to the best practice suggested in the literature [
30,
54]. In the present study, differences in both BEE and FM% between groups did not reach significance, suggesting that limb muscle asymmetry due to amputation generates similar effects to those observed in PAs-SCI with different upper and lower limb PhA. However, the limited sample size prevented us from subgrouping athletes for type and level of injury (i.e., PAs with AKA–BKA, unilateral–bilateral AMP, and paraplegia versus tetraplegia). Moreover, the selected population recruited, veteran PAs from the GSPD [
30], could reduce the generalization of the results to other PAs. From that, further studies are needed to give specific and definitive recommendations.