LP is an invasive and often uncomfortable procedure that can be technically challenging. Post-dural puncture headache (PDPH) is reported in up to 38% of cases  though other complications are uncommon .
The incidence of PDPH can be reduced by replacing the stylet before withdrawing the needle  or with the use of an atraumatic Sprotte  or Whitacre  needle.
It is generally recommended that CSF opening pressure be measured for patients with sudden onset headache . This is principally to diagnose cerebral venous sinus thrombosis (CVST) or idiopathic intracranial hypertension (IIH) if elevated and to detect low pressure headache if reduced. Measurement requires the use of larger needles (potentially increasing the likelihood of PDPH) to allow rapid pressure transduction and necessitates the patient being in the lateral position.
CVST should certainly be considered as a cause of severe headache, especially in people with a prothrombotic risk factor . However, it is much rarer than SAH and is an unusual cause of thunderclap headache. If suspected after a normal CT brain, CT venography or MRI would be a more helpful test.
There is no evidence that the practice of enforcing a period of bed rest post LP has any bearing on the development of PDPH .
It is important to understand the problems that can arise during CSF sampling, transport and interpretation in order to make a correct clinical judgement.
Traumatic LP taps occur regularly  and often mandate further investigations that may be unnecessary .
Unfortunately, none of the methods of distinguishing a traumatic tap from SAH are 100% reliable  though there is general consensus that CSF should always be assessed for the presence of xanthochromia [11,49,51]. Xanthochromia refers to the yellow discolouration of the CSF supernatant seen with haemoglobin breakdown to methaemoglobin, oxyhaemoglobin and bilirubin. Oxyhaemoglobin may develop in 2-3 hours but can occur as a result of a traumatic tap  whereas the appearance of bilirubin takes longer but only occurs in vivo.
For this reason, traditional teaching has held that LP should not be performed before 12 hours from the index episode of headache [62-64]. Samples taken earlier may not reliably demonstrate the presence of bilirubin and hence be falsely negative.
Patients in whom the diagnosis of SAH is considered but in whom the CT is normal must subsequently undergo an LP at least 12 hrs after the onset of symptoms .
Consequently, in the UK, clear national guidance (NEQAS) has been issued on how CSF specimens should be handled and analysed . The LP should ideally be performed by an experienced practitioner, the last sample protected from the light (to prevent in vitro degradation of bilirubin to deoxyhaemoglobin) and analysed in the lab as soon as possible, ideally within the hour, avoiding pneumatic delivery systems that can cause haemolysis .