One health care center’s early intervention program has resulted in a decrease in total assisted ventilation hours and improved identification of ARDS patients

Noninvasive positive pressure ventilation (NPPV) provides a means to effectively decrease Paco2 in hypercapnic respiratory failure and to increase Pao₂ in hypoxemic respiratory failure. Bilevel positive airway pressure ventilation consists of inspiratory positive airway pressure and expiratory positive airway pressure. The gradient between these two pressures results in pressure support ventilation. Utilization of NPPV at the early stages of patients’ declining conditions has some obvious advantages, including minimization of Fio₂, required lung recruitment, and prevention of atelectasis.

In spring 2002, the respiratory department at Cox Medical Centers, Springfield, Mo, felt there was an opportunity for improved outcomes in patients whose fraction of inspired oxygen (Fio₂) was sequentially increased with a primary goal of maintaining an oxygen saturation per pulse oximetry (Spo₂) of >92%, as the underlying clinical condition continued to deteriorate. We began to evaluate how we could capture these at-risk patients more quickly and, if indicated, utilize NPPV more effectively. A decision was made to develop an “NPPV Early Intervention” program.

We identified three essential conditions that needed to be met:
1. consensus of the respiratory department management
2. support and guidance from our five co-medical director pulmonologists
3. support and approval for equipment and training expenditures from our administration

A “defined” trigger was needed to activate a clinically justified early intervention assessment. Department managers, in conjunction with medical directors, agreed that reaching an Fio₂ of 0.50 was an appropriate threshold for early intervention assessment; a supervisor must now be notified to trigger the assessment when that level is reached.

Staff training was a priority and has been one of our program’s major strengths. We developed all-day training seminars for our staff members. Emphasis was placed on the clinical significance of a 0.50 Fio₂ threshold and the utilization of Pao₂/Fio₂ (P/F) ratios. As part of the training, graphs for Spo₂ and P/F ratios were presented (see Table 1).

Table 1. Charts of pulse-oximetry values and ratios of Pao₂ to fraction of inspired oxygen that were presented for staff training in early intervention assessment.

The blue area on the Spo₂ chart indicates an Spo₂ of 90% with a Pao² of 60 mm Hg on a Fio₂ of 0.50. We posed the question: “If you had a patient on the floor with these parameters, would this data raise any red flags?” Most agreed that at that time the answer would be “no.” On the P/F ratio chart, a value of <200, (shaded in brown) is one indication of adult respiratory distress syndrome (ARDS) as defined by the American European Consensus Conference.¹

Note also the blue highlighted area on the Pao₂/Fio₂ chart represents a patient with a Pao₂ of 60 mm Hg on an Fio₂ of 0.50. This is the same patient highlighted in blue on the Spo₂ chart. This patient meets the Pao₂/Fio₂ criteria for ARDS. Prior to implementation of our new standard, this patient’s Fio₂ would have been incrementally increased to 0.80, 0.90, or 1.0 while in pursuit of an adequate Spo₂. Once oxygen therapy was no longer effective, and the patient was in significant distress, an urgent intubation decision would have been required. By providing our staff with supportive data for our threshold of a 0.50 Fio₂ and obtaining their support for assessing the patient, the department was able to avoid this pitfall.

It was of major importance in our staff training to address clinical application strategies. Education was accomplished through both lecture and laboratory. There were four clinical application strategies:

•titrating initial settings rapidly,
•anticipating therapeutic goals,
•using maximal pressure levels, and
•taking predefined steps in the event of an inadequate response to NPPV.

In 2004, after 18 months of early intervention experience, a retrospective clinical study reviewing 256 NPPV patients from December 2003 to February 2004 was completed to evaluate early and aggressive application of NPPV in 85 hypoxemic failure patients (Table 2).

The ARDS intubation and mortality data were consistent with data from other studies.² Of 16 ARDS patients, only four had a diagnosis of ARDS in their chart. The data demonstrated that patients meeting the criteria for ARDS had a high mortality rate when treated with NPPV. Based on our results, the department created a process to allow RTs to assist physicians in identifying ARDS patients quickly, as well as seek their intubation within 2 to 4 hours aggressively.

The department wanted its staff to be able to identify hypoxemic versus hypercapnic failure quickly and to initiate specific strategies for each.

It was decided that an algorithm (see Figure) would be the best teaching tool to help the department achieve this goal.

The algorithm is divided into two therapeutic branches, one for the hypercapnic failure and one for the hypoxemic failure. When a patient is identified as being in hypercapnic failure and care follows the hypercapnic branch, the patient shows a rapid improvement in Spo₂ and Fio₂ can generally be decreased to approximately 0.35 Fio₂ in the first hour of NPPV. The level of hypoxemia in the hypercapnic failure patient is a result of hypoventilation rather than intrinsic compliance and diffusion pathologies.

A pH <7.25 for greater than 2 hours greatly increases the risk of NPPV failure³ and consistently higher intubation rates.⁴ If the pH is <7.25, NPPV with a PS level of 15 cm H₂O is initiated, and if the patient has signs of CO₂ narcosis, a set rate of 10-12 bpm is often used.

When a patient meets the criteria for the hypoxemic-failure branch of the algorithm, it is not uncommon, during the initial use of NPPV, to see the Spo₂ make little improvement or decrease as the EPAP or CPAP is rapidly titrated to higher pressures. Lung units are hypo-
perfused due to the rerouting of blood flow away from atelectatic alveoli.⁵ As recruitment takes place with EPAP or CPAP pressures of up to 15 cm H2O, there is now a ventilation-perfusion mismatch in the form of dead-space ventilation. When perfusion returns to the recruited alveoli, oxygen saturation improves and weaning from supplemental oxygen can begin.

When hypoxemic failure is identified, EPAP is increased to 15 cm H₂O. It is then imperative to obtain a radiograph to determine if there is a focal infiltrate and, if so, to obtain an order and initiate intrapulmonary percussive ventilation treatments. If bilateral infiltrates are present, the next step is to rule out left heart failure based on three criteria that indicate its absence: ejection fraction of >50%; B-type natriuretic peptide of <100 pg/mL; pulmonary capillary wedge pressure of <18 mm Hg.

If left heart failure has been ruled out, the RT can assist the physician in determining whether to make a diagnosis of ARDS, defined by the American European Consensus Conference6 as acute onset, a Pao₂/Fio₂ ratio of <200, and the presence of bilateral infiltrates on a chest radiograph without evidence of left-heart failure. Once the diagnosis of ARDS has been confirmed, if significant improvement is not seen within 4 hours despite the use of maximum NPPV settings, an intubation order is aggressively sought.

The department believes our patients, our department, and our organization have benefited greatly from the early intervention NPPV program. Between fiscal years 2002 and 2004, we documented an increase in NPPV hours of 141% and a decrease in invasive mechanical ventilation hours of 34%. Total assisted-ventilation hours decreased by 9,000 hours (8%). During this same period, there was an increase in patient admissions of 5%.

In October 2005, our department in conjunction with nursing service developed a rapid response team. The early intervention program was already serving this purpose to a large degree, and building it into our critical assessment team was considered optimal for our patients. In 2006, the department is looking forward to completing a new prospective comparison study on hypoxemic failure following the introduction of the NPPV algorithm strategies.

Jack Edge, RRT, is shift supervisor and Lana Shaw, RRT, is ABG supervisor, Cox Medical Centers, Springfield, Mo.

The authors would like to thank David Tucker, RRT, department director, for creating the P/F ratio and Spo2 charts.

References
1. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med. 1994;149(3 Pt 1):818-24.
2. Antonelli M, Conti G, Moro L, et al. Predictors of failure of noninvasive positive pressure ventilation in patients with acute hypoxemic respiratory failure: a multi-center study. Intensive Care Med. 2001;27(11):1718-28.
3. Hoo GW, Hakimian N, Santiago SM. Hypercapnic respiratory failure in COPD patients. Chest. 2000;117(1):169-77.
4. Confalonieri M, Garuti G, Cattaruzza MS, et al. A chart of failure risk for noninvasive ventilation in patients with COPD exacerbation. Eur Respir J. 2005;25(6):348-55.
5. West JG. Respiratory Pathophysiology—The Essentials. 3rd ed. Baltimore: Williams & Wilkins; 1985:43.
6. Monchi M, Bellenfant F, Cariou A, et al. Early predictive factors of survival in the acute respiratory distress syndrome. A multivariate analysis. Am J Respir Crit Care Med. 1998;158(4):1076-81.