One of the important features of our baby’s death was something called a sub-aponeurotic haemorrhage. On the face of it, this is a relatively rare complication of vacuum assisted delivery - an invaluable technique in difficult births for delivering babies. As far as the hospital where our baby was born is concerned, this haemorrhage is just one of those unfortunate things that can happen from time to time, an occupational hazard, and there isn’t much more than is already done, that could be done to avoid causing one.

But is that actually true?

Medical Background.

A SubGaleal Haemorrhage, or sub-aponeurotic haemorrhage, (SGH) is a life-threatening haemorrhage that develops in the scalp as a result of a head trauma. Assisted delivery is an obviously risky time for developing this kind of injury, because the baby’s head comes under enormous stress as it is pulled out. If the blood supply to the scalp is damaged, the haemorrhage can spread, usually rapidly, through the large unrestricted subgaleal tissue plane under the skin on the outside of the skull. This is potentially a very large space in a newborn, with volumes of 200 - 250 ml. To put it into context, this is virtually all the blood in a normal neonate’s body. A SGH can therefore result in severe hypovolemia (low blood volume) and up to 25% of babies who require intensive care for this condition will die. Anywhere else in the body, a haemorrhage this size would be immediately obvious, but the large surface area of the head helps to conceal it, as does general ignorance of the condition. Despite warnings of its severity by various authorities and the increasing use of vacuum assisted delivery, many health care workers are thought to have limited knowledge of it and it goes significantly under-reported.

With SGH, blood losses may be massive before the hypovolemic shock becomes evident. The medical challenge is to recognize the loss of blood, electrolytes, fluid and clotting agents before hypotension (dropping blood pressure and low circulation) sets in and treat it aggressively with fluid/blood and coagulation factors. Massive, urgent blood products may be required to maintain circulation. Pressure wrapping the head has been shown to have little impact and may even be dangerous.

Optimizing outcome depends on careful monitoring, early diagnosis and prompt treatment. Monitoring should include a minimum of 8 hours’ observation for all babies following a difficult ventouse or forceps delivery, including at least hourly recording of vital signs, regardless of Apgar scores or the need for resuscitation. A SGH presents as a boggy, fluctuant swelling of the scalp that crosses suture lines and fontanels coupled with evidence of hypovolemic shock, anemia, persistent metabolic acidosis, hyperbilirubinemia, respiratory distress, disseminated intravascular coagulopathy (DIC) and seizures.  Mortality is associated with severe hypovolemic shock and coagulopathy, particularly if, as is the case in many fatal head injuries, it’s associated with low blood oxygen saturations, where massive consumption of clotting factors in the SGH further exaggerates blood loss leading to more hypovolemia, hypoxia and acidosis.

Risks.

An odds ratio (OR) is a very useful thing that you often see in the medical literature. It compares the odds of something happening under one set of conditions relative to another set of conditions. For example, the OR of lung cancer amongst smokers relative to non-smokers is thought to be about 9, so a smoker is about 9 times more at risk of lung cancer than a non-smoker. The absolute risk might be low, but smokers are significantly more at risk than non-smokers.

Although a SGH is normally a rare complication of an assisted delivery (about 1 in 300), other factors affect this basic risk to a very large degree. For instance, the additional risks associated with a prolonged second stage of labour (OR ~7.8), fetal hypoxia (OR ~6.15) and the sequential use of instruments (OR ~2.8) are not insignificant.  In fact, taken together, the combination of delays, fetal distress and the sequential use of instruments in theatre, all of which occurred in our case, increased the absolute risk of our baby sustaining a SGH from 1 in 300 to about 1 in 2.23.

Clearly, this is a massive, not insignificant increase in the risk of SGH. Given numbers like this, it would be not at all surprising to find that virtually all cases of neonatal SGH would be preventable if;

• attempted assisted delivery occurred before the second stage of labour gets too prolonged,
• the CTG was always properly looked at for signs of distress before attempting assisted delivery (particularly by the registrars and consultants actually involved in making the decision to attempt an assisted delivery),
• and the sequential use of instruments was not attempted at all when there were any risk factors for a SGH present.

So What?

The point is that an assisted delivery can be made much more risky than it need be, and understanding the numbers can give some idea of just how much and what to do about it. The three things above are not difficult lessons to learn and would probably be much less expensive to put into practice than the consequences of not doing so.

I think these things are pretty obvious, but it seems that the hospital is not so sure . . rather than account for the haemorrhage in terms of a poorly managed labour, it preferred to blame the haemorrhage on a bleeding diathesis (a predisposition to haemorrhage) associated with an infection, because this was at least ‘natural causes’. This infection risk also has a high OR (of ~8.19, at least for intraventricular haemorrhages in premature babies with early signs of sepsis), which would have increased the risks in our case even further, making this life-threatening haemorrhage an almost certain outcome of an assisted delivery.

Related posts:

• Professionals’ views about safety in maternity services.
• The UK Policy for Group B Strep Prevention.
• Make timelines that illustrate the whole story.
• Group B Strep at our local hospital.
• Funding gap puts maternity reform at risk.