Dissertation committee:
Μαντάς Δημήτριος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Σύψας Νικόλαος, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Σαμάρκος Μιχαήλ, Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Ψυχογυιού Μίνα, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Μαντζουράνη Μαρίνα, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Πουλάκου Γαρυφαλιά, Αναπληρώτρια Καθηγήτρια, Ιατρική Σχολή, ΕΚΠΑ
Τσουρούφλης Γεράσιμος, Αναπληρωτής Καθηγητής, Ιατρική Σχολή, ΕΚΠΑ
Summary:
Introduction
Coronary artery by-pass grafting and Cardiac valve replacement surgery result in significant benefits for patients, in terms of survival and improvement of quality of life. Surgical-site infections, particularly the deep ones such as mediastinitis are challenging to treat and are accompanied with important increase of morbidity, mortality and cost of healthcare delivery. This is why although surgical wound of cardiothoracic surgeries characterized as ‘clean’ surgical antimicrobial prophylaxis or perioperative antibiotic prophylaxis (PAP) is always recommended. Perioperative antibiotic prophylaxis is a part of recommendations for prevention of surgical-site infections, in order to attain the optimal outcomes in patients undergoing surgery.
The adherence to recommendations of correct use of PAP is essential to be effective. The correct use is assessed according to specific criteria, the right selection of drug in the right dose, the right time to initiate the administration, the perseverance of therapeutic drug levels with correct redosing when is needed and the right duration. Except the increase of surgical-site infections, noncompliance with recommendations increases the Clostridium difficile infections and other negative results such as toxicity and allergic reactions. Moreover, noncompliance favors the evolution of antimicrobial resistance, especially the prolonged duration. For the all the above reasons, compliance with recommendations consists a necessary part to all antibiotic stewardship programs. The effort to improve compliance in correct use of PAP is critical, particularly in Greece due to the fact that the rate of the antimicrobial consumption and the prevalence of multi-drug resistant organisms are among the highest in Europe. Although guidelines for PAP have been issued, limited data is available regarding the degree of compliance with these guidelines in cardiothoracic patients in Greece.
Aim
The main aim of this study was the assessment of current practices of surgical prophylaxis for patients undergoing coronary artery by-pass grafting and heart valve replacement surgeries and the improvement of adherence to local guidelines of perioperative antibiotic prophylaxis. Other aims were the audit of overall antimicrobial consumption and the incidence of surgical-site infections and the incidence of other postsurgical infections.
Methods
A two-phase prospective study with a pre-post design was conducted from 2014 until 2017, in the Department of Cardiothoracic Surgery of the Evaggelismos General Hospital, with periodic surveillance, for 18 monthly random chosen periods (7 months in PRE-phase and 11 months in POST phase). All patients underwent coronary-artery by-pass grafting and heart valve replacement were included in the study without exclusion criteria. We used a specific form in order to collect data..
We collected pre-surgical and post-surgical data, patient data, medical history, operation data, antimicrobial consumption data, infection and causative pathogens data. Regarding to PAP, we collected data to assess the compliance with the specific criteria, correct choice of antimicrobial, correct time of initiation of administration, correct intraoperative re-dosing and correct duration according to local guidelines. Except PAP we collected data for the overall antimicrobial consumption (pre-surgery for 24h before operation and post- surgery until the discharge of patients, during re-hospitalization in case of re-admission for 1 year after surgery). Also, we collect data for antimicrobial consumption and we calculated DDDs. Regarding with outcomes, we assessed the length of postsurgical hospital stay, the discharge or morbity, the manifestation of surgical wound bleeding during the first 24 hours, the manifestation of renal failure (increase of creatinine with double value post than pre-phase) in the first 7 post- surgical days and the readmission for any reason until 1 year after surgery. Also, we collect data to assess the following end-points, postsurgical infections, permanent stroke (cerebrovascular accident), and re-operation for any reason for 1 year after surgery. Regarding to post-surgical infections, we collect data of surgical-site infections, bloodstream infections, respiratory tract infections, urinary tract infections, Clostridium difficile infections. Also, we classified surgical-site infections as primary (sternal wound) or secondary -site (leg/donor site) infections. We collect data for all causative pathogens and their profile of antimicrobial resistance in specific antimicrobials for all postsurgical infections. In order to collect data, we use modified form.
The intervention consisted of an extra step on the antimicrobial prescribing through the hospital information system. The prescribing physician initially had to select whether the antimicrobial would be therapeutic or prophylactic. When the physician selected ‘prophylactic’ a reminder with a link to the PAP Hospital guidelines appeared. The intervention was not restrictive regarding the choice of antibiotic, however the amount of antimicrobial dispensed was adjusted to the appropriate duration (48h for cardiac surgery). The intervention was presented to the medical and nursing staff of the hospital before the run-in period, along with training sessions regarding PAP.
Results
During the study period 1080 cardiac patients were enrolled (400 in the pre- and 680 in the post phase) with a mean age of 68.5years (SD=11.0). Most of patients were male (74.7%, 807/1080). There have been 1312 procedures, because 19% (213/1080) of patients underwent combined procedures during a single operative session. Most operative procedures were elective 69.5% (751/1080) and CABG 53.1% (574/1080). Adherence to individual elements of PAP increased, to appropriate agent [PRE 4.0% (16/400) vs POST 35% (238/680, Chi-square p<0.001], to appropriate time [PRE 93.3% (373/400) vs POST 94.9% (645/680, p=0.27ns)]. Intraoperative redosing did not administer to patients for both phases of the study. Adherence to correct management of protective levels of PAP for the 1st antimicrobial was high for both phases [PRE 99% (396/400 vs POST 99,4% (676/680), p=0,300] for the second antimicrobial was low but better in PRE-phase [PRE 12,5% (47/377 vs POST 4,1% (27/659), p=0,001]. Adherence to appropriate duration not beyond 48h increased [PRE 4.0% (16/399) vs POST 15.4% (105/680), Chi-square p<0.001]. Compliance with suggested duration was higher in simple CABG or HVAR procedures than in composite procedures and increased among surgeons [Mann Whitney test p<0.007]. The Interrupted Time Series analysis showed a significant change in the slope (0.49 to 2.75, difference 2.26, 95% CI 0.49-4.036, Supremum Wald test, p=0.02) after the intervention and evaluated the results. Pre-surgical antimicrobial administration decreased significantly after the intervention [PRE-33.3% (132/400), POST 7.8% (53/680), Chi- square p<0.001]. Administration of other than PAP antimicrobials in postsurgical period did not differ between phases, [PRE 41.5% (166/400) vs POST 39.4% (268/680), chi-square p=0.49]. The incidence of surgical-site infections increased significantly after the intervention [PRE-2.8% (11/400) vs 5.9% (40/680). Chi- square p< 0.019], primary-site were higher than secondary in POST phase [PRE-phase 1.0% (4/400), POST phase 3.8% (26/680], and secondary-site were higher than primary in PRE-phase [PRE 1.8% (7/400) vs POST 2.1% (14/680)]. Deep surgical-site infections also increased [PRE 1.6% (6/400) vs POST 4.3% (29/680)]. Other postsurgical infections also increased [PRE 6.0% (24/400) vs POST 10.3% (70/680), Chi-square p<0.016], respiratory tract infections [PRE 1.6% (6/400) vs POST 3.5% (24/680)], bloodstream infections [PRE 1.6% (6/400) vs POST 2.5% (17/680)], urinary tract infections [PRE 1.0% (1/400) vs POST 0.9% (6/680)] and Clostridium difficile infections [PRE 0.0% (0/400) vs POST 0.3% (2/680)]. Alarmingly the levels of antimicrobial resistant pathogens were gradually increased. Length of index hospitalization decreased marginally after the intervention [median, IQR 10(9-13) vs 10 8-13), p=0.001] and was significantly longer in patients with SSIs [median, IQR 12 (10-18) vs 10 (8-13), p=0.0001]. Renal failure with double value of creatinine post-surgery did not differ between phases [PRE 4.8% (19/400) vs POST 5.0% (34/680), p=0.846], similarly the readmission [PRE 10.9% (42/400) vs POST 14.1% (91/680), p=0.143], reoperation for any reason after 1 year than the initial operation [PRE 6.0% (24/400) vs POST 7.2% (49/680), p=0.498], permanent stroke [PRE 0.5% (2/400) vs POST 1.2% (91/680), p=0.143], in hospital mortality [ PRE 3.8% (15/400) vs POST 4.8%(33/680), Chi-square p=0.39] did not differ between patients with or without SSI [3.9% (2/51) vs 4.5% (46/10290].
Conclusions
Our intervention improved the compliance in correct administration of PAP. Moreover, the timing of preoperative dose of PAP was appropriate in the vast majority of cases, while further improvements are needed to ensure appropriate re-dosing. The overall compliance with all elements of PAP improved but remained low. In conclusion our intervention at the electronic prescribing system showed that improvement in compliance is attainable, although corrective interventions are needed to reinforce further the results. Considering the high level of antimicrobial resistance in Greece the optimization of PAP is of great importance.
