Post-Exertional Malaise and Pacing: when exertion triggers a crash

A committee of the National Academies in the USA, then the Institute of Medicine, proposed new diagnostic criteria for ME/CFS in 2015 after reviewing the evidence.

Under these criteria, the diagnosis requires three core symptoms: a substantial reduction in previous activities for more than six months with fatigue that is not substantially relieved by rest, together with PEM and unrefreshing sleep. In addition, there must be cognitive impairment or orthostatic intolerance, and the committee even proposed the name systemic exertion intolerance disease.

For you this means: since 2015, PEM is no longer a peripheral symptom but a core feature of the diagnosis.

A team led by Chu surveyed 150 people with chronic fatigue syndrome about the triggers, onset and duration of their PEM.

129 respondents, according to the study 90 percent of those who answered, experienced PEM after physical and mental exertion and after emotional stress. Only 11 percent reported a fixed delay of at least 24 hours, and 84 percent experienced PEM for 24 hours or longer.

For you this means: a draining conversation can count as much as a walk, and a varying onset does not argue against PEM.

A research group led by Stussman conducted nine focus groups with 43 people with ME/CFS at the US National Institutes of Health (NIH).

Of 18 participants who were interviewed after an exercise test, 17 reported that their symptoms began within 24 hours and peaked within 72 hours. The core symptoms that emerged were fatigue, cognitive difficulties and neuromuscular complaints.

For you this means: if the worst day is only the third one, the trigger easily slips out of view.

A team led by Snell had 51 women with chronic fatigue syndrome and 10 sedentary control women complete a maximal exercise test twice, 24 hours apart.

On the first day there were no significant differences; on the second, the women with CFS reached markedly lower values for oxygen uptake and work output. The study team itself writes that the first test alone would rather have suggested deconditioning.

For you this means: on a single test day you can look like someone who is merely out of shape. The difference shows in the recovery, although the control group was very small.

A research group led by Charlton compared healthy people after 60 days of strict bed rest with people with long COVID and ME/CFS and with healthy controls.

Bed rest led to muscle wasting without changing fiber types, whereas both patient groups had more fast-fatiguing glycolytic fibers. The team concludes that physical inactivity alone cannot explain the reduced exercise capacity and the muscle changes.

For you this means: according to these data, the sentence "you are just out of shape" falls short.

A team led by Hawk compared 15 people each with chronic fatigue syndrome, with major depression and without illness.

The best distinguishing features were PEM, unrefreshing sleep and problems with memory and concentration. Using symptom severity, 91.1 percent of cases were classified correctly.

For you this means: how pronounced PEM is can help with the distinction. However, the study is very small, and both illnesses can exist at the same time.

A team led by Lim pooled two-day studies in ME/CFS in a meta-analysis.

In ME/CFS, the mean values of all measures were lower on the second day, while in the controls they rose; the clearest difference was in work output at the ventilatory threshold. The team calls for further studies to distinguish ME/CFS from other illnesses with fatigue, including depression.

For you this means: the early studies point to a decline on the second day, but the differentiation remains open.

A team led by Keller tested 84 people with ME/CFS and 71 sedentary controls on two days 24 hours apart, plus 55 pairs matched for aerobic capacity, age and sex.

In ME/CFS, peak work rate fell by 5.5 percent and time to maximal exertion by 6.6 percent on the second day, among other measures, while in the controls only carbon dioxide output fell, by 3 percent. In the matched pairs, peak oxygen uptake in ME/CFS fell by 6 percent and oxygen uptake at the threshold by 6.7 percent, while both remained unchanged in the controls.

For you this means: even with the same baseline fitness, people with ME/CFS recovered less well, with declines averaging in the single digits.

A team led by Joseph compared 160 people with ME/CFS from an analysis of 1,516 invasive tests with 36 controls.

Peak oxygen uptake was 80 versus 101.4 percent of the predicted value, and 31 percent of skin biopsies were consistent with small nerve fiber damage. The team describes reduced return of blood to the heart and impaired oxygen extraction in the tissue as biologically plausible contributors.

For you this means: these are mechanisms from a selected clinic population, not a proven cause.

A team led by Singh examined 10 people with persistent exercise intolerance after mild COVID-19 and 10 matched controls with an invasive exercise test.

Peak oxygen uptake was 70 versus 131 percent of the predicted value and oxygen extraction 0.49 versus 0.78, although the pumping output of the heart was similar.

For you this means: here the limitation appears to lie in the periphery, in a very small study without PEM as an explicit inclusion criterion.

A team led by Cotler tested five questions on the duration of PEM from the DePaul Symptom Questionnaire.

They correctly classified people with ME or CFS in 81.7 percent of cases, and misclassified people with multiple sclerosis or post-polio syndrome as ME or CFS in only 16.6 percent.

For you this means: structured questions can help with the distinction, but they do not replace a medical examination.

A research group led by Appelman deliberately triggered PEM in 25 people with long COVID and PEM using a maximal exercise test and examined blood and muscle tissue before and afterwards, compared with 21 fully recovered controls.

The muscles had a lower capacity for oxygen-based energy production in the mitochondria and more fast-fatiguing fibers. After exertion, metabolic disturbances and focal tissue damage increased. Amyloid-containing deposits were already more frequent than in controls beforehand and increased after exertion in both groups, but they were not located inside the capillaries. There were no blocked capillaries and no altered capillary density.

For you this means: after overexertion, measurable changes appeared, which makes it hard to dismiss PEM as imagination. However, the study is small, and the team considers it possible that the first biopsy influenced the second.

A team led by Haffke measured the responsiveness of the inner vessel lining in 30 people with post-COVID syndrome with fatigue and exertion intolerance, 14 of them with ME/CFS, and in 15 healthy people.

It was reduced in 5 of 14 with ME/CFS and in 5 of 16 without ME/CFS criteria, and in none of the healthy people, and the vasoconstricting messenger endothelin-1 was higher in both patient groups than in the control groups.

For you this means: in some of those examined, the vessels responded measurably worse, not in all of them.

A team led by White randomized 641 people with chronic fatigue syndrome according to the Oxford criteria to specialist medical care alone or supplemented with adaptive pacing therapy, CBT or GET, with self-reported outcomes after 52 weeks.

Compared with specialist medical care alone, fatigue was 3.4 points lower with CBT and 3.2 points lower with GET, and physical function was 7.1 points higher with CBT and 9.4 points higher with GET; with adaptive pacing therapy, neither differed. Serious adverse reactions were recorded in 1 to 2 percent in each group, and the team concluded that CBT and GET could moderately improve outcomes.

For you this means: the largest treatment trial saw moderate benefits. Its validity was later discussed intensively, and the adaptive pacing therapy tested there is not the same as today's energy management.

A team led by Larun pooled eight randomized trials with 1,518 participants for Cochrane.

According to the review, exercise therapy probably reduces fatigue at the end of treatment, with moderate certainty of evidence, while long-term effects and the risk of serious adverse reactions remain very uncertain. All trials used the 1994 CDC criteria or the Oxford criteria, and the review notes that people diagnosed according to other criteria might respond differently.

For you this means: Cochrane sees a short-term benefit on questionnaires but cannot reliably assess harms. According to a 2024 editorial note, a project to update the review was discontinued.

A team led by Geraghty analysed a patient survey with 1,428 participants in comparison with other surveys.

8 to 35 percent reported a benefit from CBT, 54 to 74 percent reported clearly negative reactions to GET, and pacing was most often rated as useful, by 44 to 82 percent.

For you this means: these are important experiences, but surveys are prone to self-selection and do not replace controlled trials.

On behalf of the German Federal Ministry of Health, the Institute for Quality and Efficiency in Health Care (IQWiG) reviewed the state of knowledge on ME/CFS, in which PEM is considered a core symptom.

For GET, two studies in mild to moderate ME/CFS showed statistically significant but on average only small benefits compared with standard therapy, without it being possible to rule out a disadvantage from serious adverse events. In its overall weighing, IQWiG therefore no longer derived a hint of benefit for GET and considered a reliable weighing of benefit and harm currently not possible. For CBT in mild to moderate ME/CFS, the institute saw a hint, the weakest level of certainty, of short and medium-term benefit, but not for a longer period. Suitable studies were lacking for severe ME/CFS, and on pacing it described a glaring lack of meaningful studies.

For you this means: the German assessment is nuanced too: small measured benefits, but no established benefit of GET because harms cannot be ruled out, and large gaps.

A team led by O'Connor grouped people with ME/CFS by available energy and by whether they lived below, within or above their energy envelope.

More available energy was associated with a better level of functioning, but with overexertion this advantage was much smaller, and overexertion weighed especially heavily precisely when available energy was higher.

For you this means: overstepping can be costly precisely on better days, although a cross-sectional study does not prove causes.

In an international online survey, a team led by Clague-Baker asked 488 people with ME/CFS about pacing with heart rate monitoring.

The team notes that there is only a single study on this, even though several guidelines recommend the method. Participants named more than 30 advantages and more than 30 disadvantages and reported that heart rate monitoring had reduced the severity and duration of their PEM.

For you this means: many people experience the heart rate monitor as useful, it has hardly been tested, and the information is based on self-report.

A team led by Ghali analysed the records of 197 adults with ME/CFS at the University Hospital of Angers in France.

In 27 of 197, or 13.7 percent, new or atypical symptoms preceded a deterioration, most often mood disturbances, and in these people PEM was less intense.

For you this means: if you know your warning signs, you might be able to slow down earlier. That is a hypothesis from medical records, not a tested method.

A team led by Casson pooled 14 randomized trials on pacing programmes, alone or combined with other approaches.

Compared with no treatment or usual care, the programmes were associated with less fatigue, less psychological distress and better physical function. However, the largest improvements were seen in programmes that included a gradual increase in physical and mental activities.

For you this means: what was called pacing there is not identical to the energy management without a progression goal that NICE and the D-A-CH consensus describe.

A team led by Parker supported people with post-COVID syndrome for six weeks with a structured pacing protocol and weekly phone calls.

Among the 31 who completed the protocol, the average number of post-exertional deteriorations fell from 3.4 in the first week to 1.1 in the sixth week.

For you this means: crashes became less frequent in this small group, but without a control group it remains open what was due to pacing and what to the natural course.

Through support groups and social media, a team led by Davis surveyed 3,762 people from 56 countries with confirmed or suspected COVID-19.

After the sixth month, fatigue, PEM and cognitive dysfunction were the most common symptoms, and 85.9 percent experienced relapses, triggered mainly by exercise, physical or mental activity and stress.

For you this means: relapses after activity were the rule in this group, in a sample that is not representative.

In RECOVER, a team led by Vernon applied the 2015 IOM criteria to participants who were assessed at least six months after infection and had not had ME/CFS beforehand.

4.5 percent of the infected, 531 of 11,785, met the criteria, compared with 0.6 percent, 9 of 1,439, without infection. For the rate of new cases in participants followed from the time of infection compared with matched uninfected participants, the team reports a hazard ratio of 4.93 with a confidence interval of 3.62 to 6.71. PEM was the most common ME/CFS symptom among the infected, at 24.0 percent, or 2,830 of 11,785.

For you this means: about one in four infected people in RECOVER reported PEM. The diagnosis there is based on self-reported symptoms, which can fluctuate.

A team led by Pouliopoulou pooled studies on the frequency of PEM in adults with post-COVID condition and studies on rehabilitation.

In ten studies with 4,076 participants, the frequency of PEM was 25 percent, with low certainty of evidence. Five studies found less frequent PEM episodes after individually tailored rehabilitation with pacing elements, and seven found no indication of deterioration due to the exercise component, all with a high to very high risk of bias, and the team emphasises a subgroup without PEM.

For you this means: not all people with long COVID have PEM. For people without PEM, supervised, adapted exercise may be possible without a setback, based on limited evidence.

On 21 December 2023, the German Federal Joint Committee (G-BA) adopted a directive for coordinated, interprofessional care in suspected long COVID or a condition with a similar cause or presentation. It came into force on 9 May 2024 and has been able to take effect in care since 1 January 2025.

A medical contact person, usually the general practitioner, coordinates examinations as well as treatment and support options and manages the involvement of specialist groups and other health professions, up to specialised outpatient clinics for the most severely ill. The G-BA topic page explicitly names ME/CFS in its title.

For you this means: in Germany, the first route usually leads through a general practice. That alone says nothing about waiting times and capacity.

A team led by Froehlich surveyed 499 people in Germany online who met the Canadian Consensus Criteria and reported PEM lasting 14 hours or longer.

Participants often cited geographical and financial reasons why they could not use existing services, and reported low satisfaction with their medical care. The team concludes that there is medical undersupply and, in 2021, before the long COVID data, cites an estimate of more than 300,000 people affected in Germany. In 2023, IQWiG arrived at roughly 140,000 to 310,000 for the period before the pandemic.

For you this means: if you feel lost in the system, that is a problem of care provision and not a personal failure.

A team led by Wormgoor analysed two online surveys from Norway on 788 rehabilitation stays, 86 consultations and 89 hospital treatments in ME/CFS.

PEM was taken into account in only 48 percent of rehabilitation stays. When it was taken into account, 40.1 percent reported a deterioration of their health; when it was not, the figure was 63.2 percent, and a focus on PEM was associated with higher satisfaction.

For you this means: whether PEM was taken seriously was measurably linked to the course, in retrospective self-reports that show an association and not proof of cause and effect.