HRV After COVID: What the Research Actually Shows (and How to Rebuild Without Crashing)

Almost everyone who got COVID and wears a smartwatch noticed the same thing: their HRV fell, their resting heart rate climbed, and neither went back to normal as fast as they expected. For some people it resolved in ten days. For others it took months, and a smaller group is still waiting.

That gap between expectation and reality is where a lot of anxiety lives. So it is worth separating three things that get muddled together: what the research actually shows, what deconditioning alone can do to your numbers, and what you can reasonably do about it. The honest answer is more reassuring than the scary version and less tidy than the optimistic one.

TL;DR: A 2026 meta-analysis of 11 studies and 1,162 people found a trend toward lower HRV in people with post-COVID condition compared with healthy controls — but the effects were small, not statistically significant for any single measure, and 9 of the 11 studies had a high risk of bias. In other words, the signal is plausible but weaker than headlines suggest. Most of the early drop after an infection is a normal acute stress response, and much of the slow recovery afterwards is deconditioning from reduced activity, not permanent damage. Rebuild with graded activity, weeks not days, and stop rules based on symptoms — not on chasing your old HRV number.

YouTube: Related video

First, the short version of what HRV is doing

Heart rate variability is the variation in the time between heartbeats. A healthy heart does not beat like a metronome; it speeds up slightly on the in-breath and slows on the out-breath, and that beat-to-beat variation is largely a readout of how much influence your parasympathetic (rest-and-digest) nervous system has over your heart.

Higher HRV generally means more parasympathetic influence, which usually corresponds to better recovery. Lower HRV generally means more sympathetic (fight-or-flight) influence, which shows up with infection, alcohol, poor sleep, hard training, stress and dehydration.

Most wearables report either rMSSD (root mean square of successive differences) or SDNN (standard deviation of normal-to-normal intervals). Apple Watch reports SDNN, sampled during background readings and Breathe sessions, which is why it can vary a lot from day to day. Neither is a diagnostic tool. Both are useful as trends.

What the research actually found

The most rigorous recent synthesis is a 2026 systematic review and meta-analysis published in the European Journal of Neurology. The authors pooled 11 non-randomised studies covering 1,162 participants — 593 people with post-COVID condition, defined as symptoms more than three months after infection, and 565 healthy controls.

The findings, stated precisely, because precision matters here:

All three point in the expected direction — lower variability, more sympathetic dominance — and none reach statistical significance. Heterogeneity was high (I² of 83% for SDNN and 78% for rMSSD), and nine of the 11 included studies were rated at high risk of bias.

The authors' conclusion is appropriately careful: the meta-analysis "suggests a possible association" between post-COVID condition and altered autonomic function. Not proof of a mechanism. Not a reason to panic about a single low reading.

That does not mean nothing is happening. It means the average effect across a messy set of studies is modest, and the variation between individuals is large. Some people clearly do have persistent autonomic symptoms — palpitations, exercise intolerance, postural dizziness — and HRV is one of several places that shows up. Others have a low-ish number for a few weeks and are back to normal by spring.

There is a second, less discussed explanation for the post-illness HRV dip: deconditioning. Even a week of reduced activity lowers blood volume and stroke volume, and the body compensates with a higher resting heart rate and lower variability. Two weeks of bed rest is enough to produce measurable changes. So part of what looks like "COVID changed my HRV" is really "I did much less for three weeks than I normally do".

Why your numbers look different: four plausible reasons

1. The acute infection response. During and immediately after an infection, sympathetic tone rises and HRV falls. This is normal and temporary. It is the same pattern you would see with flu or a bad cold.

2. Deconditioning. Less movement means lower blood volume, lower stroke volume and a compensatory higher heart rate. HRV usually follows. Rebuilding takes weeks of consistent, easy activity — not a heroic return to training.

3. Sleep disruption. Illness wrecks sleep architecture. Alcohol-based cough syrups, late-night fever, disrupted routines and anxiety all suppress HRV independently of the virus.

4. Autonomic dysfunction in a subset of people. This is the mechanism the research above is probing. It appears to be real for some people with persistent post-COVID symptoms, and the practical implication is that pushing through fatigue tends to make things worse rather than better.

For most people reading this, reasons 1 through 3 dominate, and time plus graded activity fixes them. For a minority, reason 4 matters, and the approach is slower and more careful.

What a normal recovery curve looks like

This is a rough shape, not a promise. Individual variation is enormous, and the ranges below are descriptive rather than clinical.

Two things to watch for in that table. First, fitness coming back slower than HRV is normal — deconditioning takes longer to reverse than autonomic stress takes to settle. Second, a flat HRV that never returns to baseline over three months is worth a conversation with a doctor, not a reason to buy more supplements.

The trap: treating HRV as a permission slip

Once you have a number, it is tempting to let it decide everything. HRV is 62, so today is a green light. HRV is 48, so today is a write-off. This is a bad way to use the metric at the best of times, and a genuinely harmful way to use it while recovering from illness.

The reason is measurement noise. Apple Watch SDNN readings are short samples taken at irregular times. A single reading can swing 20% based on how you slept, whether you had coffee, and how long you sat still before the reading. Building a training decision on one sample is like judging the weather from one photo.

Use instead:

• The seven-day average. This is the trend line. One bad day inside a good week is noise.
• Resting heart rate. More stable than HRV, and a reliable early warning for illness and accumulated fatigue.
• Respiratory rate during sleep. A rise of 1–2 breaths per minute above your normal, sustained for two nights, is a strong signal that something is going on.
• Symptoms. The most important input, and the one wearables cannot measure.

If you want the mechanics of reading these together, our guide to using HRV and resting heart rate to decide whether to train covers the decision logic, and what an early illness signal looks like covers the acute phase.

A graded rebuild protocol

The principle is simple: add load in small steps, and let symptoms — not enthusiasm — decide when to step up. This is the same logic NICE uses in its guideline on managing the long-term effects of COVID-19, which recommends energy management and explicitly does not recommend graded exercise therapy as a treatment for post-COVID symptoms.

The distinction matters. Graded exercise therapy pushes towards a predefined target regardless of how you feel. Pacing stays under your symptom threshold and adjusts downwards when you overshoot.

The 24-hour rule is the heart of it: if a session leaves you more fatigued the next day than the day before, the session was too much. Not "you are weak". Too much. Reduce it by a third and repeat that step until it is tolerated.

Stop rules: when to back off immediately

Stop the session and rest if any of these happen:

• Post-exertional malaise — disproportionate fatigue, flu-like feelings, or a symptom flare 12–48 hours after activity. This is the signature of post-viral fatigue and the clearest signal to reduce load.
• Heart rate that will not settle — still 15+ bpm above your normal resting rate hours after finishing.
• Breathlessness at a level of effort that used to be easy, especially if it does not resolve within minutes.
• Dizziness or lightheadedness on standing. Worth mentioning to a doctor if it is recurrent.
• Sleep getting worse for more than a few nights while you are increasing activity.

None of these mean you have done permanent damage. They mean the current step is above your tolerance, and the correct response is to go back one step rather than push through.

When to see a doctor

This post is not medical advice, and some symptoms deserve an actual clinician rather than a blog post:

• Chest pain, pressure or tightness, especially on exertion
• Palpitations that are new, sustained, or associated with dizziness or fainting
• Breathlessness at rest, or oxygen saturation persistently below your normal
• A resting heart rate that stays 10+ bpm above baseline for more than a week without an obvious cause
• Exercise tolerance that is still clearly reduced three months after infection

Most people who had COVID do not need this list. But if you are in it, the sequence is a GP or physician conversation, not another wearable.

What actually helps (and what is probably a waste of money)

Things with reasonable evidence behind them:

• Sleep. Seven to nine hours, consistent timing. This is the highest-leverage variable for HRV, full stop.
• Graded activity within tolerance. The single best-supported intervention for post-viral fatigue, provided it stays under the symptom threshold.
• Adequate protein and hydration. Recovery from illness involves rebuilding tissue; under-eating slows it.
• Treating the treatable. Iron deficiency, thyroid problems and sleep apnoea all lower HRV and all have real treatments. If fatigue persists, these are worth testing rather than guessing.
• Time. Unfashionable, but most of the recovery happens here.

Things to be sceptical of:

• Supplements marketed for "HRV support". The evidence for most is thin to nonexistent, and none of them substitute for sleep and pacing.
• Cold plunges and intense sauna protocols during recovery. Both are physiological stressors. They are not what a depleted system needs.
• Pushing through fatigue to "rebuild toughness". With post-viral symptoms, this reliably makes things worse.
• Obsessing over a single daily number. See above.

If you are an athlete

Athletes tend to have a harder time with this, because the identity is wrapped up in training volume. Two things help.

First, separate "fitness" from "form". You will lose some fitness. It comes back faster than you fear — detraining is real but largely reversible, and the early losses are mostly plasma volume and glycogen, not muscle.

Second, use your watch data as a constraint rather than a target. Set a rule before you start: if resting heart rate is up more than 5 bpm and HRV is below your seven-day baseline for two consecutive days, the next session becomes easy or rest. Write it down. Decide it once, when you are calm, rather than every morning when you are impatient.

Our guide to low HRV when you feel fine covers the case where the numbers and the body disagree, which happens a lot during a rebuild.

Summary

The evidence on HRV after COVID is real but modest: a 2026 meta-analysis found lower HRV in people with post-COVID condition, with small effect sizes and wide confidence intervals. That is enough to take persistent symptoms seriously and not enough to conclude that a low reading means lasting damage.

Practically, most of the early dip is the normal infection response, and much of the slow return is deconditioning. The fix is unglamorous: sleep, graded movement, adequate food, and stop rules based on symptoms rather than on the number on your wrist.

Century AI shows your HRV, resting heart rate, respiratory rate and sleep as a trend rather than a daily verdict, which is the only way these numbers are actually useful during a rebuild. If you want the fundamentals first, start with what HRV actually means.

Give it weeks. Go back a step when you need to. The number follows the behaviour, not the other way around.

Century AI helps you understand your body with a daily health score, recovery score, and sleep insights — using the watch you already wear.