--

The American Joumal of Surgery 192 (2006) 806-811

Papers presented

A multidisciplinary clinical pathway decreases rib fracture-associatedinfectious morbidity and mortality in high-risk trauma patients

s. Rob Todd, M.D.a,*, Michael M. McNally, M.D.b, John B. Holcomb, M.D.C,Rosemary A. Kozar, M.D., Ph.D.b, Lillian S. Kao, M.D.b, Emest A. Gonzalez, M.D.b,Christine S. Cocanour, M.Df, Gary A. Vercruysse, M.D.g, Marjorie H. Lygas, M.S.d,

Bobbie K. Brasseaux, R.R.T.e, Frederick A. Moore, M.D.aaDepartment of Surgery. The Methodist Hospital. 6550 Fannin Street. Smith Tower 1661, Houston, TX 77030, USA

bDepartment of Surgery, The University of Texas Medical School at Houston. Houston, TX, USA

"Trauma Division, United States Army 1nstitute of Surgical Research. Fort Sam Houston. TX, USA

dTrauma Program, Memorial Hermann Hospital-Texas Medical Center, Houston. TX. USA

eRespiratory Care and Pulmonary Diagnostic Services, Memorial Hermann Hospital-Texas Medical Center, Houston, TX, USAfDepartment of Surgery, Division of Trauma, UC Davis Medical Center, Sacramento, CA. USA

gDepartment of Surgery. Emory University School of Medicine. Atlanta. GA, USA

Manuscript received April 15, 2006; revised manuscript August 10,2006

Presented at the 58th Annual Meeting of the Southwestern Surgical Congress, Kauai, Hawaii, April 3-7, 2006

Abstract

Background: We initiated a multidisciplinary clinical pathway targeting patients greater than 45 years ofage with more than 4 rib fractures. The purpose of the current study was to evaluate the effect of thispathway on infectious morbidity and mortality.Methods: This was a prospective cohort study. Data evaluated included patient demographics, injurycharacteristics, pain management details, lengths of stay, morbidity, and mortality. Univariate and multi-variate analyses were perforrned using a significance levei of P < .05.Results: When adjusting for age, injury severity score, and number of rib fractures, the clinical pathwaywas associated with decreased intensive care unit length of stay by 2.4 days (95% confidence interval [CI]-4.3, -0.52 days, P = .01) hospitallength of stay by 3.7 days (95% CI -7.1, -0.42 days, P = .02),pneumonias (odds ratio [OR] 0.12, 95% CI 0.04 to 0.34, P < .001), and mortality (OR 0.37,95% CI 0.13to 1.03, P = .06).Conclusions: Implementation of a rib fracture multidisciplinary clinical pathway decreased mechanicalventilator-dependent days, lengths of stay, infectious morbidity, and mortality. @ 2006 Excerpta MedicaInc. Ali rights reserved.

Keywords: Mortality; Multidisciplinary cIinical pathway; Pain management; Pneumonia; Rib fractures

Thoracic trauma accounts for 10% to 15% of trauma ad-missions and 25% of traumatic deaths [1,2]. In this popu-lation, rib fractures are present in greater than two thirds ofpatients [2,3]. Recent studies have demonstrated the impactof rib fractures after blunt trauma in the elderly population[4,5]. Bulger et aI documented that patients older than 65years of age with more than 4 rib fractures were more likely

* Corresponding author. Te!.: + 1-713-441-6149; fax: + 1-713-790-6472.

E-maU address: srtodd@tmh.tmc.edu

to die or suffer serious setbacks when compared to thoseaged 15 to 64 [5].

Although elderly trauma patients are a vulnerable group,multiple studies have documented that age-related morbid-ity increases before age 65 in the trauma population [6-11].Easter proposed that patients as young as 40 years of ageexhibit increased morbidity [12]. With this knowledge inhand, we previously studied our population to determine therelationship between age and number of rib fractures inblunt trauma patients [13]. We demonstrated that patientsolder than age 45 who had more than 4 rib fractures were atincreased risk for prolonged mechanical ventilator depen-

0002-9610/06/$ - see front matter @ 2006 Excerpta Medica Inc. Ali rights reserved.doi: 10.10 16/j .amjsurg.2006.08.048

107

S.R. Todd et aI. / The American Journal of Surgery /92 (2006) 806-8//

, I

dent days, shock trauma intensive care unit length of stay,and overalI hospital length of stay. Secondary to thesefindings, we instituted an aggressive rib fracture multidis-ciplinary cIinical pathway to target this at-risk population.The purpose of the current study was to evaluate the effectof this pathway on infectious morbidity and mortality. Wehypothesized that implementation would decrease infec-tious morbidity and mortality.

MethodsThis was a prospective observational cohort study per-

formed at Memorial Hermann Hospital in Houston, TX,which is the lead regional Levei I Trauma Center forTrauma Service Area Q, which incIudes nine counties in theupper Gulf Coast of Texas with a population of approxi-mately 4.5 million. Based on our previous retrospectivestudy, we instituted a multidisciplinary cIinical pathway inFebruary 2002 targeting patients greater than 45 years ofage with more than 4 rib fractures. Patients with concurrentsevere traumatic brain injury, defined as a Glasgow ComaScale score :58 and an abnormal brain computed tomogra-phy scan, were admitted to the Neurosurgery Trauma In-tensive Care Unit and thus were not candidates for thestudy.

Patients meeting the aforementioned criteria were admit-ted to a monitored bed (Surgical Intermediate Care Unit orShock Trauma Intensive Care Unit) where they receivedpatient-controlIed analgesia and incentive spirometry uponadmission. On the first 3 days of hospital admission, thebedside nurse evaluated the patients on three criteria: pain,inspiratory volume, and cough. Pain was assessed duringincentive spirometry or coughing using a visual analoguescale (score from 1 to 10) with failure being a score greaterthann 6. Inspiratory volume was determined using the in-centive spirometer. A volume less than 15 mL/kg was con-sidered failure [14]. For the third criteria, the bedside nurseperformed a subjective assessment of the patient's cough,with a weak cough being considered failure. Patients whofailed 1 or more of these criteria on any of the first 3 daysof hospital admission were entered into the multidisci-plinary cIinical pathway.

Fig. 1 depicts the multidisciplinary pathway. EnrolImentmandated focused efforts from the respiratory therapy, pain,physical therapy, and nutrition services. The respiratorytherapists implemented the trauma service's volume expan-sion protocol, which utilizes aerosolizedpharmacologicther-apies, the EzPAP positive airway pressure system (SmithsMedical, St Paul, MN), and other escalating invasive ther-apies. The pain service was pivotal in maximizing paincontroI. They prescribed oral pain medications (incIudingnonsteroidal anti-inftammatory medications), intravenouspain medications, and epidural analgesia. The goal of thephysical therapists was to optimize mobility via strengthen-ing, range of motion, and balance exercises. The nutritionservice monitored the nutritional status of the patient andoptimized it utilizing varying diets and supplements. Oncean enrolIed participant passed alI 3 of the entrance criteriaon 3 consecutive days, he or she was removed from themultidisciplinary cIinical pathway. EnrolIees who no longerrequired Shock Trauma Intensive Care Unit or Surgical

807

Intermediate Care Unit levei of care prior to being removedfrom the pathway were transferred to the ward with ahigh-risk respiratory care consult.

A prospective cohort of 150 patients since the pathway'sinception in February 2002 served as the study group (post-pathway patients). For comparison, a historic control of150 patients prior to the pathway's inception served as thecontrol group (pre-pathway patients). Participants in bothgroups were older than 45 years of age with more than 4 ribfractures.

Data were obtained from the Trauma Registry, pharmacydatabase, infection control continuous quality improve-ment database, and electronic medical recordoThe TraumaRegistry was reviewed for patient demographics (age, race,gender, and mechanismof injury),injury characteristics(chestabbreviated injury scale scores, injury severity scores, num-ber of rib fractures, and the presence of ftail chests, stemalfractures, pulmonary contusions, hemothoraces, pneumo-thoraces, combination hemo/pneumothoraces, and thoracos-tomy tubes) and mortality. Details of the pain managementregimen, including the utilization of nonsteroidal anti-in-ftammatory drugs, patient-controlIed analgesia pumps, andepidural catheters, were obtained from the pharmacy data-base. The infection control continuous quality improvementdatabase was queried to identify pneumonias (using stan-dard Centers for Disease Control and Prevention defini-tions), mechanical ventilator dependent days, shock traumaintensive care unit length of stay, and total hospitallength ofstay. The electronic medical record was used to confirm theaforementioned data.

In comparing the 2 study groups, the Student t test wasused for continuous data. If the assumptions of this test werenot met, the appropriate test (Mann-Whitney U test) wasapplied. A chi-square analysis was used for categorical data.When appropriate, this was substituted with the Fisher exacttestoUnivariate and multivariate (linear and logistic regres-sion) analyses were performed to determine associationsbetween the utilization of the multidisciplinary cIinicalpathway and mechanical ventilator-dependent days, lengthsof stay, infectious morbidity, and mortality. Numeric dataare presented as means :t SD and median (interquartilerange) where appropriate. The odds ratios (ORs) and the95% confidence intervals (CIs) were caIculated. A P valueless than .05 was considered significant. Number CruncherStatistical Systems (Kaysville, UT), 2004 was used for alIstatistical analyses.

The colIection and review of data was approved by TheUniversity of Texas Health Sciences Center at HoustonCommittee for the Protection of Human Subjects and theMemorial Hermann Hospital Institutional Review Board.

ResultsA prospective cohort of 150 patients from February 2002

to October 2004 served as the study group and was com-pared to a historic control of 150 patients from February1999 to February 2002. The study patients compared to thehistoric controls were younger [56 (51-65) years vs. 60.5(52-72) years, P = .02] but similar in terms of their gender[94 males (63%) vs. 97 males (65%), P = .72].

Table 1 displays the injury characteristics of the 2groups. The study group was similar in terms of their chest

108

808 S.R. Todd et aI. / The American Joumal of Surgery 192 (2006) 806-811

no

no

yes

no

Rib Fracture PathwayAssessment

On momlng rounds assess

1. ISvolume2. PainScale (1.10)(dynamic)3. CoughScore

Weak=1Strong =2

Rlb Fracture Cllnlcal Pathway

1. Pain consult for pain managemenl optimizalion(pain service pager 713-608-xxxx)

2. RT consull for Volume Expansion Protocol3. PT/OT consull (improve palient mobility)4. Nutrition consult (assessment and improvemenl)5. Nurse Practitlonerassessment and evaluationof

co-morbid factors during momlng rounds6. Low moIecular weight anti-coagulants musl nol be

administered prior 10paln ecnsul!. If they have beenstarted, they must be held 24 haurs prior to epidural

IS, Pain, and Cough will be assessed and recorded q day untilpalient passes ali three crilaria for Ihree consecutive days or untildischarge trom SIMUlSTICU

Patient is not eligible for pathway

patlent responslveness definedIn non-Intubated patients as GCS >= 13In Intubated patlents as GCS components eyes =4 and motor Ia6

Fig. I. Rib fraclure multidisciplinary clinical pathway. SIMU = Surgical Intermediate Care Unit; STICU = Shock Trauma Intensive Care Unit; IS

incentive spirometry; RT = respiratory therapy; PT = physical therapy; OT = occupational therapy; GCS = Glasgow Coma Scale.

abbreviated injury scale score [4 (3-4) vs. 4 (3-4), P = .17]and injury severity score [21 (17-29) vs. 21 (17-29), P =.67] in comparison to the historic controls. Despite this, thestudy patients had more rib fractures, fiail chests, hemotho-races, pneumothoraces, and hemo/pneumothoraces.

Overall, the pain management intervention altered pa-tient care significantly. Post-pathway patients were pre-

scribed patient-controlled analgesia significantly more thanpre-pathway patients [138 (92%) vs. 47 (31%), P < .0001].Similarly, post-pathway patients witnessed a significant in-crease in epidural catheter utilization [63 (42%) vs. 17(11%), P < .0001]. An additional16 (11%) post-pathwaypatients were offered or had an unsuccessful attempt atepidural catheter analgesia.

109

S.R. Todd et aI. I The American Journal of Surgery /92 (2006) 806-8//

Table 1

Injury characteristics of the two study groups

No. of rib fractures

Flail chest

Stemal fracture

Pulmonary contusionHemothorax

Pneumothorax

Hemo/pneumothorax

Thoracostomy tube

Control group(pre-pathway

patients)(n = 150)

6 (5-7)7 (5%)8(5%)

55 (37%)22 (15%)58 (39%)21 (14%)75 (50%)

<.0001<.0001

.13

.55<.0001

.01<.002

.01

Study group(post-pathway

patients)(n = 150)

7 (6--9)41 (27%)15 (10%)50 (33%)65 (43%)79 (53%)43 (29%)54 (36%)

P value

Table 2 depicts the unadjusted univariate analyses whencomparing the outcome data of the 2 study groups. Post-pathwaypatientshad nonsignificantlydecreased shock traumaintensive care unit and hospitallengths of stay, yet statisti-cally significant decreased pneumonia and mortality rates.When adjusting for age, injury severity score, and numberof rib fractures, the clinical pathway was associated withdecreased shock trauma intensive care unit length of stay by2.4 days (95% CI -4.3, -0.52 days, P = .01), hospitallength of stay by 3.7 days (95% CI -7.1, -0.42 days, P =.02), incidence of pneumonia (aR 0.12, 95% CI 0.04 to0.34, P < .001), and mortality (aR 0.37, 95% CI 0.13 to1.03, P = .06).

CommentsThe true incidence of rib fractures following trauma is

unknown and difficult to quantify [15]. However, recentstudies suggest that 10% of trauma center admissions pos-sess rib fractures and rib fractures are considered a markerof severe associated injuries [4,16-19]. Dependent on theinjured rib, the associated injuries may involve the heart,great vessels, liver, ancllor spIeen. Fractured ribs are aIsofrequently associated with underlying thoracic trauma (i.e.,pulmonary contusions, hemothoraces, pneumothoraces, andcombination hemo/pneumothoraces). These injuries con-tribute to the pulmonary morbidity seen in multi-systemtrauma.

Recent studies have demonstrated the impact of rib frac-tures after blunt trauma in the elderly population [4,5].Bergeron et ai documented that despite lower indices ofinjury severity, mortality was significantly increased in pa-tients greater than 65 years of age with rib fractures [4].They concluded that elderly patients with rib fracturesshould receive special attention. Although trauma patients"':~~~~":.:.U\':n~.: ~k~';~'~~~~'~~\ n~h.:: ""~~::~~~:...~..:;:-:~~~~~\,,\e~~~~~~~~t.:~: ..:!t:::'~~~~7:~~h.:~~'~~'~~\~.~~nY-~r:~~~:~\~'t~h'~~;;~~;:,.{lfnálr;.~,l!/I!~~~~.~~t::~f{1Jz(!~i~~~~lÇ!_if~~!Ú~(;~~:lIi'~~:;~~!f-~;!{k~;;~:~l:}lft~>-_

809

Our current study focuses on the effectivenessof this path-way. The study group had similar indices of injury to thecontrol group as evidenced by their chest abbreviated injuryscale scores and injury severity scores. The high injuryseverity scores represent rib fracture patients as severelyinjured trauma patients. Similarly, Bergeron et ai docu-mented a mean injury severity score of 19.9 among theirpatients, and Bulger et ai found a mean injury severity scoreof 21 [4,5]. Likewise, the high abbreviated injury scalescores depict the severity of the thoracic trauma in the studycohort. The 1990 revision of the abbreviated injury scalescores defines a 4 as representing a severe "threat to life" fora given injury in a particular body region [20].

Despite the similar injury indices, the post-pathway pa-tients had statistically significant more rib fractures. Theassociated injuries due to these rib fractures included ftailchests, sternal fractures, pulmonary contusions, hemothora-ces, pneumothoraces, and combination hemo/pneumothora-ces. Ali ofthese were increased in the post-pathway patientsexceptfor sternaIfracturesand pulmonarycontusions.Trunkeyestimated that 50% of rib fractures cannot be detected byplain anteroposterior chest roentgenograms [21]. Through-out the course of this study, we increased our utilization ofchest computed tomography scans in the initial evaluationof blunt trauma patients. We believe that this change inpractice may have accounted for our increased diagnoses ofrib fractures and their associated thoracic injuries.

On unadjusted univariate analyses, the post-pathway pa-tients had non-significant decreased shock trauma intensivecare unit and hospital lengths of stay, yet a statisticallysignificant decreased incidence of pneumonia and mortality.When adjusting for age, injury severity score, and numberof rib fractures, the clinical pathway was associated withdecreased shock trauma intensive care unit length of stay,hospitallength of stay, incidence of pneumonia, and mor-tality.

Enrollment in the rib fracture multidisciplinary clinicalpathway mandated focused efforts from the respiratory mer-apy service, pain service, physical therapy service, andnutrition service. Of these, the aggressive utilization of thepain service was the one care arm lacking prior to pathwayinception. Pain service consultation altered care plans sig-nificantly. The post-pathway patients were prescribed pa-tient-controlled analgesiaand epidural catheter analgesia sig-nificantly more than the pre-pathway patients. We believethis aggressive pain control was critical to the success of our

Table2Unadjusted univariate analyses of the outcome data of the two studygroups

'i>~uó.":I~<>U\> P 'I~\\l~l\><>..~-\>"",...~~.r.:uh...l':.:..

.

810 S.R. Todd et aI. / The American Joumal of Surgery 192 (2006) 806-811

pathway. This is supported in the literature [17,18,22-26].Wisner et aI documented that epidural analgesia decreasedpulmonary complications and mortality in elderly traumapatients [22]. Similarly, multiple studies have reported com-parable outcome measure improvementsto include decreasedmechanical ventilator-dependent days, lengths of stay, pul-monary complications, and mortality [17,18,23-26].

Ours is not the tirst rib fracture multidisciplinary clinicalpathway reported in the literature. In 2001, Easter docu-mented a similar pathway aimed at optimizing the mobili-zation, respiratory care, and pain management of rib frac-ture patients [27]. Similarly, the success of our pathwayimplementation has been corroborated by Sesperez et aI[28]. They concluded that clinical pathways improve patientoutcomes in tive key trauma conditions including rib frac-tures.

In conclusion, increased morbidity has been proposed intrauma patients as young as 40 years of age [12]. In ourprevious study, we demonstrated that patients older than 45years of age who had more than 4 rib fractures were atincreased risk for prolonged mechanical ventilator-depen-dent days, shock trauma intensive care unit length of stay,and overall hospitallength of stay [13]. Secondary to thesetindings, we instituted a rib fracture multidisciplinary clin-ical pathway to target this at risk population. In this study,we document that implementation decreased mechanicalventilator dependent days, lengths of stay, infectious mor-bidity, and mortality.

References[I] Besson A, Saegessen F. Color Atlas of Chest Trauma an Associated

Injuries, Volume 1. Oradell, NJ: Medical Economic Books, 1982.[2] Shorr RM, Crittenden M, Indeck M, et aI. Blunt thoracic trauma:

analysis of 515 patients. Ann Surg 1987;206:200-5.[3] Newman RJ, Phil D, Jones IS. A prospective study of 413 consecutive

car occupants with chest injuries. J Trauma 1984;24:129-35.[4] Bergeron E, Lavoie A, Clas D, et aI. Elderly trauma patients with rib

fractures are at greater risk of death and pneumonia. J Trauma2003;54:478-85.

[5] Bulger EM, Ameson MA, Mock CN, et aI. Rib fractures in theelderly. J Trauma 2000;48: 1040-7.

[6] Tometta P, Mostafavi H, Riina J, et al. Morbidity and mortality inelderly trauma patients. J Trauma 1999;46:702-6.

[7] Demetriades D, Sava J, Alo K, et aI. OId age as a criterion for traumateam activation. J Trauma 2001;51:754-7.

[8] Perdue PW, Watts DD, Kaufmann CR, et aI. Differences in mortalitybetween elderly and younger adult trauma patients: geriatric statusincreases risk of delayed death. J Trauma 1998;45:805-10.

[9] Davis JW, Kaups KL. Base deficit in the elderly: a marker of severeinjury and death. J Trauma 1998;45:873-7.

[10] McOwin O, Melton SM, May AK, et aI. Long-term survival in theelderly after trauma. J Trauma 2000;49:470-6.

[11] Scalea TM, Simon HM, Duncan AO, et aI. Oeriatric blunt multipletrauma: improved survival with early invasive monitoring. J Trauma1990;30: 129-36.

[12] Easter A. Management of patients with multiple rib fractures. Am JCrit Care 2001;10:320-9.

[13] Holcomb 18, McMullin NR, Kozar RA, et aI. Morbidity form ribfractures increases after age 45. J Am ColI Surg 2003; 196:549-55.

[14] Pierson DJ, Kacmarek R. Foundations of Respiratory Care. NewYork: Churchill Livingstone, 1992.

[15] Quaday KA. Morbidity and mortality of rib fractures. J Trauma1995;39:617.

[16] Ziegler DW, Agarwal NN. The morbidity and mortality of rib frac-tures. J Trauma 1994;37:975-9.

[17] Wu CL, Jani ND, Perkins FM, et aI. Thoracic epidural analgesiaversus intravenous patient-controlled analgesia for the treatment ofrib fracture pain after motor vehicle crash. J Trauma 1999;47:564-7.

[18] Bollinger CT, Van Eeden SF. Treatment of multiple rib fractures:randomized controlled trial comparing ventilatory with nonventila-tory management. Chest 1990;97:943-8.

[19] Wisner DH. A stepwise logistic regression analysis of factors affect-ing morbidity and mortality after thoracic trauma: effect of epiduralanalgesia. J Trauma 1990;30:799-805.

[20] Copes WS, Sacco WJ, Champion HR, et aI. Progress in characterizinganatomic injury. In Proceedings of the 33rd Annual Meeting of theAssociationfor the Advancement of Automotive Medicine. Baltimore,MD; October 2-4, 1989;205-18.

[21] Trunkey DD. Cervicothoracic trauma. In: Blaisdell FW, Trunkey DD,editors. Trauma Management. Volume 111.New York: Thieme, 1986.

[22] Wisner DH. A stepwise logistic regression analysis of factors affect-ing morbidity and mortality after thoracic trauma: effect of epiduralanalgesia. J Trauma 1990;30:799- 805.

[23] Moon MR, Luchette FA, Oibson SW, et aI. Prospective, randomizedcomparison of epidural versus parenteral opioid analgesia in thoracictrauma. Ann Surg 1999;229:684-91.

[24] Bulger EM, Edwards T, Klotz P, et aI. Epidural analgesia improvesoutcome after multiple rib fractures. Surgery 2004;136:426-30.

[25] Mackersie RC, Karagianes TO, Hoyt DB, et aI. Prospective evalua-tion of epidural and intravenous administration of Fentanyl for paincontrol and restoration of ventilatory function following multiple ribfractures. J Trauma 1991;31:443-9.

[26] Ullman DA, Fortune 18, Oreenhouse BB, et aI. The treatment of

patients with multiple rib fractures using continuous thoracic epiduralnarcotic infusion. Reg Anesth 1989;14:43-7.

[27] Easter A. Management of patients with multiple rib fractures. Am JCrit Care 2001;10:320-9.

[28] Sesperez J, Wilson S, Jalaludin B, et aI. Trauma case managementand clinical pathways: prospective evaluation of their effect on se-lected patient outcomes in five key trauma conditions. J Trauma2001;50:643-9.

DiscussionDan Vargo, M.D. (Salt Lake City, UT): Clinical path-

ways have been shown to improve outcomes in many clin-ical situations. From the Parkland formula to ambulationguidelines after joint replacement, elimination of dogmaticvariations in care has lead to better patient care. This manu-script outlines a clinical pathway for the management ofpain after chest trauma. It utilizes age and number of ribfractures, in addition to admission to the ICU or intermedi-ate care unit as criteria for eligibility. Criteria include thingsthat document a decrease in vital capacity: IS and cough. Intheir 150 patients compared to historical controls, the chestAIS is similar. Despite this, there is a decrease in associatedmorbidity and mortality, in addition to a decrease in thenumber of chest tubes placed in the study group despitemore chest injuries. Also, there is a longer time spent onmechanical ventilation in the study group compared to thehistorical controls. The increased injury as noted is probablyexplained by the use of CT scan, but the decrease in tubethoracostomy would not be explained by this. In addition,it would not explain the increased amount of times spenton the ventilator. An outcome is also not provided forpatients who went out of the unit while still on their clinicalpathway.

So with this, I have a few questions for the authors. First,were patients ever transferred directly from the ICU to thefloor while they were still on the pathway? And if so, whatis the rate of retum for these patients? In your manuscript,

S.R. Todd et aI. / The American Joumal of Surgery /92 (2006) 806-8// 81l

you document that these patients were taken off the pathwaywhen they went to the fioor and so it would be interesting tosee that the pathway actualIy led to a decreased rate ofretum to the intensive care unit. Secondly, why were pa-tients intubated longer in the study group? You have pa-tients who have a decreased infection rate, but yet they areon the ventilator longer which we would alI love to have inour ICUs; so could it be that they benefited from having thatextra day in the ICU for either better pain management orfor clearance of secretions or so forth? Third, you show anincreased use of PCA and epidural in your study patients.Does this mean that surgical house staff or the people whoare taking care of these patients just don't understand whatpain control is and why did you need the pain service to beinvolved to show you that these patients needed to havebetter pain control? And then lastly, why would you onlyuse these criteria in patients over the age of 45? If itdocuments that you have a better outcome, I would arguethat most of my pain management problems are not greaterthan the age of 45, they are in the young healthy male 20-to 45-year-old who will sit there, and cry and whimperinstead of going through and doing any pulmonary toilet.

Patrick J. Offner, M.D., F.A.C.S. (Wheatridge, CO): Itis remarkable that patients were on the ventilator longer, yetyour pneumonia rate was decreased. And I have to wonder:was there another difference between your control groupand your study group? In other words, in the study period,did you implement any other protocols, for instance venti-lator bundles? Were you elevating the head of the bed?Were you changing your nutrition at ali? We you doingother things that just this rib fracture pathway that couldhave affected your pneumonia rate?

Roxie M. Albrecht, M.D. (Oklahoma City, OK): I haveone questionoI got a copy of your manuscript ahead of time

and I noticed that you included patients that were over theage of 45. However, in your results section, there wasnobody between the ages 45 and 51. They don't break theirribs?

S. Rob Todd, M.D. (Houston, TX): Dr. Vargo in refer-ence to the first question about bounce-backs from the fioorto the ICU. When patients went directly from the ICU to thefioor, they received a "high risk pulmonary." This meansthey receive aggressive pulmonary care by RespiratoryTherapy and mobilization by Physical Therapy. The secondquestion as far as why these patients might possibly beintubated longer, I propose that this was secondary in-creased chest trauma. When we look at these patients, theyhad more rib fractures, pulmonary contusions, hemothorax,etc. Unfortunately, by the chest AIS they get a 4, but theirISS did not refiect an increase in severity to their chesttrauma. I think they may have had worse chest trauma andrequired longer ventilation. As far as your third question,the PCA utilization, alI patients were ordered PCAs onadmission and we only used the pharmacy database toretrieve those because they had alI the data for us specifi-cally, but ali patients were supposed to have a PCA orderedon admission. Obviously 92% were done. And the fourthquestion, why don't we do this on ali of our patients? Weactually for the most part do these days. I think it probablywould be interesting to look at those patients younger than45 and see how they perform. I suspect they would make astatistical difference in their infectional morbidity and mor-tality. Dr. Offner, the biggest thing that we did differentlywe weren't doing prior to implementation of this pathwaywas aggressive pain management (specifically epidural uti-lization). Dr. Albrecht, the numbers you are referring torepresent the interquartile range, 51-65 years of age.

Reprinted with permission the American Joumal of Surgery 2006; 192(6):806-11.Todd SR, McNally MM, Holcomb JB, et aI.Amultidisciplinary clinical pathway decreases rib fracture-associated infectious morbidity and mortality in high-risk trauma patients.Copyright 2006 iÇ)EIsevier, Inc.

112