Columns, Patient Care

Study Shows Lactate Levels May Miss Sepsis

Issue 12 and Volume 41.

The Research

Boland LL, Hokanson JS, Fernstrom KM, et al. Prehospital lactate measurement by emergency medical services in patients meeting sepsis criteria. West J Emerg Med. 2016;17(5):648–655.

The Science

An EMS agency serving two large areas in and around Minneapolis attempted to determine the value of obtaining prehospital lactate levels in patients suspected of sepsis.

They used traditional criteria of sepsis, which required two or more of the following systemic inflammatory response syndrome (SIRS) criteria:

  • Heart rate ≥ 90 beats per minute;
  • Respiratory rate > 20 breaths per minute; or
  • Body temperature < 96.8 degrees F or > 100.4 degrees F along with a history or presentation suggestive of infection.

The EMTs and paramedics were provided a two-hour educational session on the signs and symptoms of sepsis along with training on the use of a handheld lactate meter, which measures lactate through a capillary blood sample from a finger stick, and temporal artery thermometer.

In a two-year period, 112 patients were enrolled. Prehospital lactate levels ranged from 0.8 to 9.8 mmol/L.

Thirteen (12%) of the 112 patients had a lactate level > 4.0, the level generally associated with sepsis, and 81% were admitted to the hospital where 24% were diagnosed with sepsis.

Patients with elevated lactate levels were more likely to be admitted to the ICU and diagnosed with sepsis as compared to those with normal lactate levels; however, this difference wasn’t statistically significant.

They concluded that this combination of sepsis detection didn’t justify its use as a method of EMS prenotification to the ED of a patient suffering from suspected sepsis.

Doc Wesley Comments

Sepsis is, without a doubt, a serious condition resulting in tens of thousands of hospitalizations and a mortality rate around 30%.

Early detection improves outcomes and therefore it’s been proposed that EMS consider sepsis detection and ED prenotification just as with other time-sensitive conditions like myocardial infarction and stroke.

However, sepsis is a sneaky condition. It has no constellation of signs, symptoms and blood tests that are uniquely diagnostic of its presence. It often presents with vague findings that require the astute clinician to consider sepsis in a large number of varying patient presentations.

In the ED, it’s not uncommon for a clinically competent physician to admit patients with suspected congestive heart failure, cardiogenic shock and run-of-the-mill pneumonia to only later find they were septic.

This study found that even in a high-volume EMS agency, the number of sepsis patients is relatively low.

The authors looked specifically at the correlation of blood lactate level and the ultimate diagnosis of sepsis. They found that this test is too sensitive, resulting in too many false positives, and poorly specific, resulting in too many false negatives.

The presence of lactate isn’t specific to sepsis. It can be increased in any condition that results in anaerobic metabolism such as hypoxia and shock from causes other than sepsis.

Two hours of education in what is one of the most challenging medical conditions may simply be insufficient to ensure EMS providers have the clinical knowledge required to detect sepsis. Additional time should be devoted to the potential value of point-of-care lactate and role capnography may play in detecting sepsis.

Medic Wesley Comments

I get excited when an agency provides training and education above and beyond the standard curriculum.

With that said, I wish this session had been longer and had been presented to all providers in the system. Giving the providers the information to allow them not only to make the assessment, but to understand the assessment findings and further communicate them to the ED staff, will help to save lives.

I also wish that there had been ED physician buy-in to work with every one of the study patients. This would have provided a better understanding of outcome variables, and would give prehospital providers the follow-up and validation of their assessment.

It’s my understanding that the most accurate lactate levels are obtained from venous blood and not capillary samples. Also, if you leave the tourniquet on the arm too long the lactate levels are falsely elevated.

Therefore, using lactate as a standalone one-time value isn’t sufficient to detect clinical conditions. But, instead, serial values should be drawn to determine trends in the patient’s lactate levels (i.e., rising or falling).

It’s clear from this study that the detection of sepsis by EMS has a way to go before it can be relied on to the same degree that other serious conditions are.