What this is
A working set of notes on why our post-lumbar puncture bed rest is now 15 minutes instead of 2 hours, what the trial evidence actually says, and what we now ask the patient or the ward to watch for.
Why The Policy Changed
For decades the standard line was “lie flat for two hours after a lumbar puncture.” It made intuitive sense. You’ve put a hole in the dura (the tough outer membrane around the spinal cord), and lying flat feels like it should give the hole a chance to close before cerebrospinal fluid (CSF, the clear fluid that bathes the brain and spinal cord) starts dribbling out under gravity.
The trouble is that when people actually tested this in trials, it turned out the bed rest didn’t do what everyone thought it did. Patients who got up straight away got headaches at the same rate, sometimes a slightly lower rate, than patients who were kept flat for hours. So we now ask the patient to lie still for about 15 minutes after the needle comes out, and that’s it. The two hours flat in bed wasn’t doing any work.
The whole reason for the change
Long bed rest doesn’t prevent the headache it was meant to prevent, and the trials have been saying that for over twenty years.
The Theory That Bed Rest Helps
The headache we’re trying to prevent is called post-dural puncture headache (PDPH). The mechanism is reasonably well understood. The needle leaves a small hole in the dura. CSF leaks out through that hole into the surrounding tissue faster than the body can replace it. The total volume of CSF in the system drops, intracranial pressure falls, and the brain effectively sags inside the skull when the patient stands up. Pain-sensitive structures (the meninges, the cranial nerves, the upper cervical nerves, and small blood vessels) get pulled on. The blood vessels also dilate as a compensatory response. The result is the classic postural headache, fine when lying down, awful when sitting or standing.
The bed rest theory was that lying flat would slow the leak. Gravity wouldn’t be pulling CSF out, the hole would seal sooner, and the patient would never become symptomatic. It’s a reasonable-sounding hypothesis. It just doesn’t hold up.
The dura is a thin sheet of tough connective tissue, and the hole the needle leaves doesn’t close because the patient is horizontal. It closes because the fibres are elastic and the tissue heals. Whether the patient is on their back or walking around the carpark seems to make no measurable difference to that process.
What The Trials Actually Show
The evidence has been remarkably consistent for a long time. The first big systematic review pulled together sixteen randomised trials, comparing patients who were mobilised straight after lumbar puncture (or rested briefly) against patients who were kept flat for anywhere from half an hour to a full day. Across more than two thousand patients, the relative risk of headache was essentially 1 in both directions, no benefit either way.
The Cochrane Collaboration updated this in 2016 and reached the same conclusion. They went further. When they restricted the analysis to only the highest-quality trials, the bed rest group actually had a slightly higher rate of PDPH than the immediate-mobilisation group. Not enough to prove that bed rest causes headaches, but more than enough to say it isn’t preventing them.
What this means in practice
The Cochrane reviewers concluded that routine bed rest after lumbar puncture should no longer be recommended as a way of preventing PDPH. There is no evidence supporting it. That’s the basis for our 15-minute policy and for most modern centres’ move in the same direction.
A more recent emergency department study looked specifically at length of stay. In their cohort, 37.5% of patients in the bed rest group developed PDPH versus 44.2% in the no-bed-rest group, again no statistical difference. But the bed rest group spent hours longer in the department, took up trolleys, and added nothing measurable to safety.
Where The Longer-Rest Position Comes From
It’s worth being fair to the older view. A few things still make people instinctively reach for “lie down longer.”
The first is that lying down genuinely helps the headache once it’s there. The headache is postural by definition, it eases when the patient is supine. So if a patient who’d been sat up early developed PDPH and felt better when they finally lay down, the obvious-but-wrong inference was that lying down earlier would have stopped it happening at all. The trials tease these two things apart and show that one (treatment) is true and the other (prevention) isn’t.
The second is that some older case series (uncontrolled, no comparison group) reported lower headache rates with bed rest. These predate the era of randomised trials and were the basis for the original 2-hour and 4-hour and 24-hour policies that grew up around the procedure. They’re not wrong about what they observed in their own units. They just couldn’t tell whether the bed rest was doing the work or whether it was something else, the needle they used, the operator, the patient population.
The third is that there’s a small, separate literature on lumbar puncture combined with intrathecal chemotherapy in haematology patients, which sometimes finds an optimal supine duration of an hour or two. That’s a different population with a different procedure (drug instillation through the needle, not just CSF sampling), and the conclusions don’t translate to a routine diagnostic LP.
The honest summary
The strongest evidence (randomised trials, meta-analysed) says bed rest doesn’t prevent PDPH. The case for longer rest comes from older uncontrolled observation and from confusing “lying down treats the headache” with “lying down prevents it.”
What Actually Reduces The Headache Rate
If bed rest doesn’t help, what does? This is the more interesting question, because it’s where the modern policy gets its safety margin. The big lever is the needle itself.
Atraumatic (pencil-point) needles, the Sprotte and Whitacre tips, the rounded ones with the side-port, almost halve the PDPH rate compared to the older cutting-tip Quincke needles. The Cochrane review on needle design puts the relative risk at about 0.4. Number-needed-to-treat to prevent one headache is roughly 14 to 24, depending on the population. This is a much bigger effect than anything bed rest was ever claimed to do.
Smaller gauge needles also help. Finer needle, smaller hole, smaller leak. The trade-off is that finer needles are slower to flow and a little fiddlier, but for diagnostic LPs the small-gauge atraumatic needle is now the default in most centres.
Bevel orientation parallel to the spine. For the cutting-tip needles, turning the bevel so it slides between the longitudinal dural fibres rather than cutting across them also reduces the leak. This is the operator’s call, but it’s worth being aware of.
What the doctor reaches for matters more than what we do afterwards
The single biggest determinant of whether the patient gets a headache is the needle that was used. A 22G atraumatic Sprotte means the patient is unlikely to have trouble regardless of what happens in recovery. A larger cutting needle in a young female patient means the headache risk is real no matter how long they lie flat.
Who’s Most Likely To Get It
Worth knowing, because it shapes how cautious to be at discharge.
Young adults between 18 and 50 are the highest-risk group. Older patients (over 60) have the lowest rates, possibly because of age-related changes in the dura and CSF dynamics. Female patients have higher rates than male patients, and pregnant patients higher again. People with a previous history of PDPH are more likely to get it next time. Low body mass index is a mild risk factor. People with a chronic headache history seem slightly more vulnerable.
These risk factors don’t change what we do, they just change how seriously to take the conversation about what to watch for. A 28-year-old woman is the patient most likely to come back two days later with a headache. A 78-year-old man almost never will.
What The Headache Looks Like
The diagnosis of PDPH rests on its postural quality. It is the textbook description, and it really does present this way.
The patient feels fine lying down. Within seconds to a couple of minutes of sitting or standing up, a dull, heavy headache starts, usually frontal or occipital, often radiating into the neck and shoulders. Lying back down resolves it within a few minutes. The pattern is reproducible and obvious.
It usually shows up between 24 and 48 hours after the procedure, but it can appear any time within the first five days. About two-thirds of headaches that are going to happen will have appeared by day two.
Associated symptoms can include neck stiffness, nausea, photophobia (light sensitivity), tinnitus (ringing in the ears), and sometimes muffled hearing. These are all from the same underlying process, low CSF pressure pulling on cranial nerves.
Worth flagging on discharge
A headache that comes on when you sit up and goes away when you lie back down is the one that fits PDPH.
A headache that’s there constantly regardless of position, or one that feels nothing like that pattern, doesn’t fit and needs different thinking.
What To Tell The Patient
The new discharge message has to do real work because the patient is leaving sooner than they used to. The conversation goes roughly like this.
Most people are fine. About 15 minutes lying still after the procedure is enough, there’s no need to stay flat for hours. They can sit up, get dressed, eat, and head off as they normally would.
A small number of people develop a headache in the next few days. It’s a particular kind of headache, comes on when standing, settles when lying down. It usually shows up in the first day or two but can appear up to five days out.
If that headache develops, the management is simple. Lie flat for a couple of hours, keep up the fluids, take simple analgesics (paracetamol, ibuprofen if not contraindicated). Most settle.
If it doesn’t settle, they need to contact the referring doctor. If they’re an outpatient and can’t get hold of the referrer, the GP or the local emergency department is the right next stop. Don’t tough it out.
The ward version of the same conversation
For inpatients going back to the ward, the message to the nursing staff is: head up, mobilise normally, watch for postural headache over the next 24 to 48 hours. If headache develops, the patient lies flat for two hours, hydrates, simple analgesia. If it persists or escalates, contact the team.
If A Headache Develops
The sequence we ask the ward or patient to follow:
-
First, two hours flat in bed. This is the treatment dose of bed rest, not the prevention dose, and it works because of the postural nature of the headache. Lying flat reduces the symptoms while the dural hole is sealing.
-
Hydration helps. Oral fluid is fine, IV if they’re inpatient and not tolerating oral. The mechanism here is partly volume support for the CSF and partly just keeping the patient comfortable.
-
Simple analgesia: paracetamol around the clock, ibuprofen if there’s no contraindication. Caffeine has a small evidence base, a strong cup of coffee or tea is sometimes worth trying.
If after that the headache hasn’t settled, or if it gets worse, that’s when the referring team needs to know. Outpatients call their referrer or, failing that, their GP, or present to ED. The team will decide whether to escalate to the next step, which in stubborn cases is an epidural blood patch, a small amount of the patient’s own blood injected into the epidural space at the puncture level by an anaesthetist, which clots over the dural hole and seals it. It’s the gold-standard treatment when conservative measures aren’t working.
Red Flags: Not A Normal PDPH
Most post-LP headaches are textbook PDPH and follow the script above. A few things break the script and need urgent review rather than another round of paracetamol:
-
A headache that’s not postural (present and unchanged whether the patient is upright or lying flat) isn’t PDPH and needs other causes considered.
-
New neurological symptoms: weakness, numbness, visual changes that aren’t just photophobia, confusion, drowsiness, fever.
-
A change in the character of the headache after several days, like a postural headache that suddenly becomes constant, or an old headache that suddenly worsens dramatically.
-
Seizures.
Any of these need imaging and a neurology or anaesthetics review. The differentials include subdural haematoma (a rare but serious complication of low CSF pressure pulling on bridging veins), cerebral venous sinus thrombosis (a clot in the brain’s drainage veins, which is also more likely after CSF leak), or meningitis (infection of the meninges). None are common, but they’re the reason the safety net exists.
The "atypical headache" rule
If the patient describes a headache that doesn’t fit the postural pattern, don’t assume it’s just a more severe version of PDPH. Different beast, different workup. The same applies to any focal neurology, fever, or change in conscious level.
Numbers Worth Knowing
| Stat | Value | Why it matters |
|---|---|---|
| PDPH risk with cutting (Quincke) needle | ~10–40% | The needle, not the bed rest, is the big lever |
| PDPH risk with atraumatic (Sprotte/Whitacre) needle | ~2–6% | Roughly halves the risk vs cutting needles |
| Cochrane review conclusion | No benefit of bed rest vs immediate mobilisation | Basis for moving to short-rest policies |
| Typical PDPH onset | 24–48 hours, up to 5 days | The discharge warning has to cover this whole window |
| Highest-risk patient profile | Young female, low BMI, prior PDPH | The patient most worth being explicit with at discharge |
| Lowest-risk patient profile | Over 60, male | Almost never get it |
| First-line treatment if headache develops | Bed rest 2 hours, fluids, simple analgesia | The “treatment” dose, not the “prevention” dose |
| Definitive treatment if conservative fails | Epidural blood patch | Anaesthetics-led, very effective |
Sources
- Posture and fluids for preventing post-dural puncture headache — Cochrane Review (Arevalo-Rodriguez 2016)
- Does bed rest after cervical or lumbar puncture prevent headache? Systematic review and meta-analysis — Thoennissen et al. (CMAJ 2001)
- Does Bed Rest Prevent Post–Lumbar Puncture Headache? — Annals of Emergency Medicine
- The Impact of Bed Rest Policy Following Lumbar Puncture on PDPH and ED Length of Stay — PMC
- Postdural Puncture Headache — StatPearls (NCBI)
- Statement on Post-Dural Puncture Headache Management — American Society of Anesthesiologists
- Consensus Practice Guidelines on Postdural Puncture Headache — Multisociety International Working Group (2024)
- Atraumatic vs Conventional Needles for Lumbar Puncture — Academic Emergency Medicine (Gottlieb 2021)
- Needle gauge and tip designs for preventing PDPH — Cochrane Review
- Incidence and risk factors of postdural puncture headache: prospective cohort study — Perioperative Medicine
- Post lumbar puncture headache: diagnosis and management — PMC
- Causal-effect of bed rest and PDPH in diagnostic LP: prospective cohort — JCMA
Last updated May 17, 2026.