A recent publication in Chest by Semler et al examined the utility of the ramped position for the emergent intubation of patients in the ICU (1). The results not only call into question our use of this technique, but more importantly the outcomes we use to evaluate the efficacy of airway interventions outside the pristine environment of the operating room.
The authors performed a multicenter pragmatic trial in which they randomized patients undergoing emergent endotracheal intubation at 4 ICUs, to be placed in either the sniffing or ramped positions during the intubation process. Patients were excluded if the necessity of a definitive airway precluded the time required for randomization, or the treating physician felt a specific position was required for the safety of the patient.
Of the 311 ICU patients intubated by the Critical Care fellows during the one year enrollment period, 260 were included in the trial. The most common reasons for intubation were sepsis, altered mental status and pneumonia. The groups appeared well balanced in baseline characteristics as well as pre-intubation difficult airway characteristics, method of pre-oxygenation, oxygen saturation at induction, induction agent and neuromuscular blocker, as well as choice of laryngoscopy performed.
Overall the patients randomized to the sniffing position had easier intubations than those in the ramped cohort. Operator-reported Cormack-Lehane grade of view and ease of intubation were better with sniffing position than ramped position. First pass success occurred more frequently in patients randomized to the sniffing position (85.4% vs 76.2%; p = .02). Fewer patients randomized to the sniffing position required the use of a bougie to secure the airway (6.2% vs 19.2%; p= 0.002) or a second laryngoscopic device (6.2% vs 16.2%; p = .01).
Despite these procedural advantages, the use of the sniffing position did not seem to improve the ability to pre-oxygenate the patient, nor did it extend the apneic period. The median lowest oxygen saturation did not differ between the two groups (92% (79-98) vs 93% (84-99%) p=0 .27). In fact, though not statistically significant, the patients randomized to be intubated in the sniffing position experienced more episodes of desaturation below 80% than those in the ramped position (28.3% vs 20.5% p=0.14).
On first glance this paper seems to be a concerning damnation for the ramped position much championed in the halls of the FOAM-verse. But before we discard our much beloved sheets and pillows let us examine a number of key subtleties.
First what did the authors mean when they utilized the sniffing and ramped position and how does this compare to the traditional uses of the terms? They cite a number of studies that compare the sniffing and ramped positions in the operating room. In a trial by Lee et al (2) the authors defined the sniffing position as:
The ramped position was defined as:
Similarly in Collins et al (3) defined the sniffing position as:
And the ramped position as:
In contrast, Semler et al defined the sniffing position as:
And the ramped position as:
These definitions are noticeably different, and in fact Semler et al’s definition of the sniffing position is fairly in line of what has traditionally been defined as a ramped position. Essentially these authors examined which method of aligning the ear to sternal notch resulted in better intubating conditions. Specifically whether the optimal position was obtained when starting from a bed flat position or first elevating the head of the bed before applying your traditional ramp. It appears at least from a procedural perspective, obtaining the ear-to-sternal-notch position from a bed flat position creates better intubating conditions.
But this of course fails to account for the greater question. What is the appropriate outcome to grade studies examining tools and techniques used in emergent endotracheal intubations? Does first past success truly incorporate all the considerations that are required to safely intubate the critically ill? Is first pass success in isolation truly a clinically meaningful outcome?
Sackles et al found a strong association with increasing number of attempts to secure a definitive airway, and the rate of adverse events (AE) observed (4). One or more AEs occurred in 14.2% of cases in the first attempt group vs 53.1% of cases in the multiple attempts group. While certainly concerning, these results are not unexpected. The large majority of these AEs were desaturation and esophageal intubation. Although these events are not preferred they are often the primary way a laryngoscopic attempt ends in lieu of a secured airway. In fact, the rates of hypotension, cardiac arrest and aspiration did not seem to rise until four or more intubation attempts were made.
And so multiple attempts may be associated with increasing rates of AE, ensuring first pass success does not directly translate into improved clinical outcomes. Despite Semler et al’s determination that the sniffing position demonstrated improved intubating conditions, and required less use of the bougie and other rescue techniques, this anatomic success did not translate into improved clinical outcomes. The authors found no difference in lowest median oxygen saturation. Furthermore, patients randomized to the sniffing position group experienced more frequent episodes of severe hypoxia and even had trends towards a greater in-patient mortality (49.2% vs 40.8% p=0 .17). While laying someone flat before aligning the external auditory meatus and the sternum may be ideal from an anatomic perspective, it is not always logistically possible or advised in the severely physiologically compromised patient requiring endotracheal intubation. Prioritizing anatomical perfection at the cost of physiological stability does not lead to better clinical outcomes. Using outcome measures that do not account for the multitude of variables that define success when intubating the critically ill will invariably lead to skewed results. In place of first pass success, or even median lowest oxygen saturation should we not be examining a more clinically important outcome, perhaps successful endotracheal intubation without hypoxia, or hypotension?