Studies were considered relevant for the review if they (1) addressed either muscle cramps or skeletal muscle spasms and (2) specifically reported cramps and/or spasms in patients with upper motor neuron disease, and (3) were published, full-text available. Studies were excluded if they referred to muscle spasms of the internal organs or facial muscles, reported cramps/spasms as an adverse effect to other treatment, referred to patients with concurrent non-neurological pathology, or were non-English language. Case reports, abstracts, and poster presentations were excluded, whereas all other study types (cohort, cross-sectional, retrospective, RCT, qualitative, review) were included in the scoping review.

MEDLINE and EMBASE databases were searched in April 2022 for relevant studies, and the search strategy and final search string were reviewed and consulted with an experienced librarian. Suspecting that “muscle cramps” and “spasms” are used interchangeably, we conducted a specific search for these keywords combining MESH-terms, abbreviations, and alternative wording:

The patient population was limited to individuals diagnosed with damage to the upper motor neurons due to either stroke, traumatic brain damage, multiple sclerosis, cerebral palsy, spinal cord injury. The appearance of both UMND and muscle cramps/spasms in the same citation was ensured using the Boolean operator “AND.” Both search strings for MEDLINE and EMBASE are attached in the Appendix.

All results (n = 5,391) were exported to Covidence. After removing duplicates (n = 1,189), the remaining results (n = 4,202) underwent abstract screening. This screening process was performed by two independent assessors with expertise in the field. In case of disagreement during the abstract screening, the full text of the study in question was read, and consensus was reached through discussion. Subsequently, all included studies (n = 435) underwent full-text review by one of the assessors. Data were extracted from studies that met the eligibility criteria (n = 247). Additionally, six additional relevant articles were identified and added to the scoping review through reference list exploration, giving a total of 253 included studies.

In Covidence, a data charting form was utilized, which was developed based on the protocol. The form was refined through an iterative pilot extraction process and thoroughly discussed by both assessors to ensure collection of reliable data. Cases of doubt were reviewed together to reach a common understanding of how to extract data. Extracted information from the studies included main characteristics (title, publication year, study design, patient group), study conclusion, and specific data on spasms/cramps (e.g., use of the term, relation to ‘spasticity’, definition, clinical characteristics, measurement methods, patient’s perceptions, relation to muscle pain, EMG characteristics, and triggering techniques).

Reckoning that ‘muscle cramps’ rarely are described in the literature on UMND, we chose, prior to the extraction process, 12 well-known features (involuntary, painful, measurable by EMG, visible, localized to specific muscles, sudden onset, duration of seconds to minutes, relief through stretching or voluntary antagonist activation, and correlation with exercise, fatigue, sleep, or a shortened state of the muscle) that are associated with cramping in healthy individuals in highly-cited reviews (1, 20). In the following, we refer to these as ‘preselected cramp features’ and use these as a basis for comparison with the descriptions of spasms in the literature on UMND. In addition, the categories “No description” and “Others” were added to the extraction template, and after the extraction process, all “Other” descriptions were analyzed and subgrouped if occurring more than once. All data were exported to Excel (.csv format) for data processing (i.e., grouping and summarizing in tables). For most of the extraction process, there was no room for interpretation, but occasionally, it was difficult to assess the relationship between spasms/cramps and spasticity. In these cases, the study was categorized as ‘others’, and we subsequently evaluated whether new subcategories could be defined. Also, errors were corrected before summarizing data, primarily due to misspellings, mistakes, or inaccurate categorization.

Half of the 253 studies (n = 129) included for full-text reading had no description of cramps nor spasms. The remaining half provided only sparse descriptions, of which most could be assigned to one of the predefined cramp feature categories (Figure 3). However, as several studies on cramps (n = 8) and spasms (n = 61) highlighted that the contractions could interfere with everyday activities, this subcategory was added to the bar chart (Figure 3). The category “others” covered the remaining descriptions (n = 62) that did not fall within the predefined groupings. These mainly included the classification of spasms as reflexes (n = 23) with a subdivision of spasms into clonus, flexor, and extensor spasms, and the triggering by various stimuli (n = 28; electrical, cold, touch, or passive movement).

Special attention was given to the relation between the involuntary muscle contractions and pain: We found that 76 (out of 216) studies referred to spasms as ‘painful’, and equivalently 4 (out of 7) studies on cramps. Very often, spasms (n = 47) and cramps (n = 6) occurred more frequently in a specific setting, such as during sleep, exercise, or fatigue. Supplementary Table S3 (Appendix) provides an overview of the different ways in which cramps/spasms qualitatively are described in relation to exercise, and fatigue.

Thirty studies used patient questionnaires to evaluate the prevalence of spasms/cramps in UMND, and the self-reported frequency varied significantly among these surveys (15 to 99%). Various methods (>25 different measures) were used to measure spasms/cramps in UMND (Figure 4). Most measures (n = 212; 70%) rely on patients’ perception of muscle spasms, e.g., through Penn Spasms Frequency Scale (PSFS), Spasm Frequency Score (SFS), Numerical Rating Scale (NRS), spasm count, or a self-made scale etc. In 5 % of these studies, in which self-reporting was used as a measure, did the participants receive brief information on the presentation/meaning of cramps/spasms prior to self-reporting.

A measurement tool designated for evaluating spasms was used in 108 studies, of which the PSFS was the most frequently used (n = 72). Forty-five studies (n = 45) applied a tool, such as Ashworth Scale (AS), H-reflex size, or NRS, that was nor invented for measuring spasms, and 38 studies (n = 38) applied electromyography to measure spasms. The remaining studies (n = 113) either did not report anything regarding how they measured spasms (n = 24) or used a very unspecific measure, e.g., undefined spasm count, scale, or subjective improvement. Figure 4 provides an overview of the numerous methods that are used to quantify spasms/cramps in the literature on UMND.

In 32 of the included studies, spasms/cramps were deliberately triggered, primarily electrically/magnetically (n = 17), or via triggers, such as warm, cold, tickling (n = 17), or stretch (n = 6). Electromyography was, in addition to verifying the presence of spasms, used to determine the stretch reflex (n = 16), H/M-ratio (n = 15), H-reflex (n = 13), cutaneomuscular reflex (n = 10), clonus (n = 8) etc., and in several studies spasms were correlated to the size of these muscular reflexes.

We conducted a systematic literature search, which aimed to explore the definition, prevalence characterization and measurement of cramps and spasms in the literature on UMND. Based on these findings, we evaluated if current clinical and laboratory measures can distinguish muscle cramps from spasms while considering shared pathophysiological mechanisms that could be inferred from the studies.

To investigate an under-recognized phenomenon poses a puzzle. Thus, reckoning that ‘muscle cramps’ rarely are described in the literature on UMND, we used 12 pre-defined, widely-recognized features associated with cramping in healthy individuals as a benchmark for comparing the descriptions of spasms in the literature on UMND.

The systematic search confirmed our initial hypothesis that ‘cramping’ is confined to specific patient groups other than UMND (1, 2, 21, 22), as illustrated in Figure 2. In addition, less than 10 % of the studies defined either cramps or spasms, and among those covering both terms (n = 29 studies), only two differentiated between them. Thus, the literature displays a lack of consensus regarding the semantics, and definition of these.

Critics will argue that this demonstrates a mere case of semantic nitpicking. However, we argue that unclear definitions along with the recognition that only half of the studies describe the muscle contraction in question (Figure 3) contribute to confounding and misleading use of terms.

Half of all studies (n = 129, 51%) lacked explicit descriptions of cramps or spasms, and the remaining studies, while predominantly using the term ‘spasms’, predominately aligned with preselected cramp features. This underscores the significant convergence in characterizing cramps and spasms (Figure 3), reaffirming their clinical resemblance.

Spasms typically persist for a few seconds to minutes, occur spontaneously as verified by EMG (23, 24), or in relation to specific triggers (e.g., touch, movement). Although we find no records on whether patients can voluntarily terminate their spasms (e.g., using stretch or activation of the antagonist), many patients report subjective improvement in spasms (25)/spasticity (26) when doing stretch. Also, spasms often develop under similar circumstances as cramps, such as during the night ‘similar to nocturnal leg cramps’ (27, 28), when subjects are fatigued, or in relation to exercise (Supplementary Table S3).

Notably, the self-reported spasm frequency varies considerably across surveys, potentially due to differing patient interpretations of ‘cramps’ and ‘spasms’, and lack of objective measures. Reviews on cramps propose a differentiation criterion based on the presence of pain in cramps (1, 2, 22). In contrast, a cross-sectional study on 61 patients with spasticity challenges this notion by remarking that the general concept of “painful” spasms is inaccurate since about two-thirds of the involuntary movements were painless (29). We find that every third study (n = 76) describe spasms as painful, underscoring a possible disagreement across the literature. This discrepancy introduces ambiguity in both scientific literature and patient accounts - and suggests the potential existence of a partially distinct, yet co-occurring, phenomenon such as cramps.

As the association to exercise is a core feature in muscle cramping in healthy individuals (30), we specifically addressed exercise and fatigue in relation to spasms in UMND (Supplementary Table S3).

Following a neurological insult, individuals tend to become less physically active, potentially leading to more fatigable muscles during everyday activities and physiological testing (31).

This perspective aligns with some patients’ perception that too much activity is associated with fatigue and a further increase in spasticity (25) and that spasms are linked with physical overexertion (29), extreme fatigue (32), uncustomary exertional efforts (33), fatigue, overuse, and inflammation (34). Similarly, cramps in healthy individuals are associated with exercise, especially with beginning of a new exercise program (22) or particularly after periods of inactivity (13).

The escalation of sensory inputs induced by physical exercise likely plays a role in this correlation, yet specific exercise interventions have demonstrated an immediate exacerbation of spasms with heightened physical exertion (32, 33). Does this imply that individuals with upper motor neuron disease (UMND) should abstain from physical exercise to mitigate the risk of spasms?

In contrast, we posit that individuals with UMND will adapt over time to a consistent training regimen, experiencing reduced muscle fatigue and fewer spasms during daily activities. Addressing this potential impediment is crucial for clinicians recommending exercise to UMND patients. The prospect of a positive, long-term impact of FES-assisted exercise on muscle spasms is suggested by the experiences of individuals with spinal cord injuries (35). Nonetheless, dedicated interventions examining the effect of training on muscle spasms are warranted (33, 34).

Confusing muscle cramps with spasms may ultimately lead to different treatments, as muscle spasms are often considered part of spasticity that arises following CNS damage. Many clinicians use a broad definition of spasticity (36) which places greater demands on specifying the individual spasticity-symptoms, as hypertonia, clonus, hyperreflexia, and spasms can exist independently and do not necessarily share common pathophysiology (37).

Unfortunately, we find that a clear distinction between spasms and spasticity is not made in the studies on UMND, among which only one-fifth (n = 57) explicitly demarcate ‘spasms/cramps’ from ‘spasticity’ as distinct symptoms. Research has explored the ambiguity between spasms and spasticity, revealing conflicting expert views on their distinction (38), as well as incongruence between clinicians’ and patients’ vocabulary and understanding of the terms (25).

We speculate that spasms and spasticity, which often go hand in hand, are treated as if it is one and the same. This may explain why clinicians seem more likely to resort to antispastic pharmaceuticals when treating muscle spasms than when treating muscle cramps.

The consequence of miscommunication is potentially significant, as the prevalence of cramps and spasms foremost (~50%) are quantified with different variations of self-reported spasm frequency tools (PSFS, SFS, NRS etc.) without aligning interpretations of cramps/spasms before the assessment. We find that there is a need to develop a better way to quantify spasms seeing that one-fifth of the studies do not report how they measure spasms, whereas the remaining use more than 25 different methods, many of these unvalidated (17, 39) and with poor correlation between self-reported spasms and (semi)objective measures (e.g., EMG (23, 40, 41), Spinal Cord Assessment Tool for Spastic Reflexes (SCATS) (14, 42), plantar stimulation response (43), H-reflex depression (44)).

Evidently, similar methodological challenges pertain to the measurement of muscle cramps and spasms, facing that both are challenging to predict and examine. Therefore, spasms/cramps are deliberately elicited through electric/magnetic stimulation or physical manipulations in 32 of the included studies. We highlight those comparable procedures (voluntary activation (45), fatigue (31), nociceptive stimulation, tendon vibration, and electrical stimulation with increasing frequency (46)) are used to provoke both spasms and cramps. Interestingly, the EMG firing response to the external stimulation shows similarities during spasms in UMND (47, 48) and cramps in healthy individuals (49, 50). This suggests a potential mechanistic overlap.

Ultimately, miscommunication and lack of objective measures may lead to false conclusions, as exemplified by studies, which make conclusions on ‘muscle spasms’ but measure another part of spasticity, such as increased muscle tone or reflex excitability [e.g., using Modified Ashworth Scale (MAS) or H-reflex size].

Therefore, we stress that experts and patients have very different perceptions of muscle cramps, spasms, spasticity, and their interrelatedness, and we raise the concern that the incidence of misdiagnosis of spasms/cramps in UMND is considerable.

Recognizing that part of the patient’s spasticity may be attributed to muscle cramps necessitates a critical reconsideration of current treatment strategies, seeing that muscle spasms usually are treated with CNS depressing medication (15, 16), whereas cramps primarily are addressed with non-pharmaceutical therapies that are largely unexplored in individuals with neurological disease (17, 18).

Thus, this recognition also emphasizes the future perspective for investigating if part of the involuntary contractions in spasticity may be mitigated using treatment strategies typically employed for muscle cramps. If so, this approach may reduce the reliance on antispastic medication and improve motor rehabilitation efforts that could otherwise be compromised by CNS-depressing medications.

The discussion of this scoping review is constrained by the fact that a search examines how the scientific literature describes muscle cramps and spasms in UMND, and thus it is possible that clinicians in practice have more nuanced understanding of these phenomena. However, it is most likely that the literature and clinical practice reflect each other.