A pilot study to determine the feasibility of continuous cefazolin infusion

A novel method for quantification of cefazolin local tissue concentration in blood, fat, synovium, and bone marrow using liquid chromatography - mass spectrometry

To be effective, the concentration of antibiotic used must exceed the minimum inhibitory concentration (MIC) against infecting organisms at and in the surgical site. Few studies follow antibiotic levels for tissues that are manipulated during surgery. The aim of this work was to develop and validate a novel LC-MS method as well as an efficient extraction technique for the quantification of cefazolin in local tissues and whole blood. This method uses the same efficient extraction method across multiple tissue types affected by orthopedic surgery: blood, fat, synovium, and bone marrow. The ability to quantify cefazolin in these tissues will help identify surgical techniques and antibiotic dosing protocols that better protect patients from infection. The internal standard, 13C2,15N-cefazolin, co-elutes with cefazolin, and was used in calibration curves and tissue extracts as well as for cefazolin recovery and matrix effects. The protocol was rigorously tested, including measurements of reproducibility and calibration curve quality. The recovery of the extraction method ranges from 94% to 113% across all sample types. There is little to no matrix effect on cefazolin signal (98-120%). The developed method was used to determine cefazolin concentrations in tissues of 10 patients undergoing a total knee replacement. Cefazolin blood concentrations were approximately 500 times higher than in adipose, synovium, and bone marrow tissues. This clinical data shows that although the minimum inhibitory concentration is largely surpassed in blood, the concentration of cefazolin in fat, synovium, and bone marrow could be insufficient during a knee replacement. This method of cefazolin quantification will help surgeons optimize antibiotic concentrations in the local tissues during knee replacement surgery and potentially reduce serious post-surgical infections.

A validated LC-MS/MS method for the quantitation of cefazolin in human adipose tissue: Application of EMR-Lipid sorbent as an efficient sample clean-up before mass spectrometric analyses

A novel, simple, rapid, and sensitive high-performance liquid chromatography-tandem mass spectrometry method was developed to determine cefazolin concentrations in human adipose tissue. Sample preparation was performed by protein precipitation followed by using Captiva EMR-Lipid plates. The mobile phase consisted of 5 mM ammonium formate and 0.1% formic acid in water and 0.1% formic acid in ACN, and was pumped through a Synergi Fusion-RP column with a gradient elution program at a flow rate of 0.3 mL/min. The mass spectrometer was operated in a positive ion mode. Cloxacillin was used as an internal standard due to the observed cross-signal contribution between cefazolin and ¹³C₂,¹⁵N-cefazolin. The method was validated according to the FDA and EMA guidelines and passed all the acceptance criteria. The calibration range was 0.05-50 µg/mL in adipose tissue homogenate (0.15-150 µg/g in adipose tissue), precision CV < 4.5%, accuracy within 93.1-100.4%. The carry-over was negligible, recovery of the method was high, and no significant matrix effect was present. Rat subcutaneous adipose tissue was demonstrated to be a suitable surrogate matrix for human adipose tissue. The validated method was successfully applied in a pilot pharmacokinetic study and will further be used in a large cohort of non-obese and obese patients dosed prophylactically with cefazolin before surgeries.

Improving intraoperative administration of surgical antimicrobial prophylaxis: a quality improvement report

Despite widespread adoption of the Surgical Care Improvement Programme, the incidence of surgical site infections (SSIs) remains high. It is possible that lapses in appropriate administration of antimicrobial prophylaxis may play a role. We noted significant discordance with national guidelines with regards to intraoperative antibiotic administration at our institution, leading to implementation of a quality improvement initiative using multidisciplinary education and reminder-based interventions to improve prescribing practices and increase compliance with national guidelines. We observed a significant improvement in adherence to all aspects of antibiotic administration guidelines as a result of such interventions. Targeted multidisciplinary interventions may help improve prescribing practices of surgical antimicrobial prophylaxis and provide an opportunity to potentially decrease the burden of SSI and the related morbidity and mortality.

Cefazolin Concentration in the Mediastinal Adipose Tissue of Patients Undergoing Cardiac Surgery

Objective

To measure the concentration of cefazolin in the anterior mediastinal adipose tissue of patients undergoing cardiac surgery, determining the variation of cefazolin concentration.

Methods

Two samples of approximately 1g of subcutaneous tissue were collected from 19 patients who underwent surgery in December 2015: the first sample was collected right after sternotomy and the second one, before sternal synthesis with steel wires. Antibiotic dosage was administered through high performance liquid chromatography.

Results

We observed a positive and statistically significant correlation between time 1 and cefazolin concentration (r=0.489 and P=0.039). For time 2 and cefazolin concentration, there was a negative and statistically significant correlation between both variables (r=-0.793 and P<0.001). A negative correlation was also observed between body mass index and cefazolin concentration at time 2 (r=-0.510 and P=0.031). The regression model showed that every 1-minute increase in time 1 corresponded to an increase of 0.240 µg/dL in cefazolin concentration, whereas every 1-minute increase in time 2 corresponded to a reduction of 0.046 µg/dL in cefazolin concentration. As for body mass index, every 1 kg/m² increase corresponded to a reduction of about 0.510 µg/dL in cefazolin concentration.

Conclusion

There was a positive and significant correlation between the initial time of surgery and cefazolin level in the first dosage. The evaluation of the second dosage showed a negative and significant correlation between cefazolin level and the second time of dosage. The concentration of cefazolin is under the influence of body mass index.

Association between Pre-Operative Cefazolin Dose and Surgical Site Infection in Obese Patients

BACKGROUND

A fixed dose of cefazolin results in serum concentrations that decrease as body mass increases. Current national guidelines suggest a pre-operative cefazolin dose of two grams may be insufficient for patients ≥120 kg; thus a three gram dose is recommended. These recommendations, however, are based on pharmacokinetic rather than outcome data. We evaluate the efficacy of pre-operative cefazolin two gram and three gram doses as measured by the rate of surgical site infection (SSI).

PATIENTS AND METHODS

We conducted a retrospective review of adult patients ≥100 kg who were prescribed cefazolin as surgical prophylaxis between September 1, 2012 and May 31, 2013 at an academic medical center. Patients were excluded if cefazolin was prescribed but not administered, had a known infection at the site of surgery, or inappropriately received cefazolin prophylaxis based on surgical indication. The SSIs were identified by documentation of SSI in the medical record or findings consistent with the standard Centers for Disease Control and Prevention definition. Inpatient and outpatient records up to 90 days post-operative were reviewed for delayed SSI.

RESULTS

Four hundred eighty-three surgical cases were identified in which pre-operative cefazolin was prescribed. Forty-seven patients were excluded leaving a total of 436 patients for final analysis: 152 in the cefazolin two gram group and 284 in the three gram group. Baseline demographics were similar between groups with a mean follow-up duration of 77 days for both groups. Unadjusted SSI rates were 7.2% and 7.4% (odds ratio [OR] 0.98, p = 0.95), for the two gram and three gram groups, respectively. When differences in follow-up between groups were considered and logistic regression was adjusted with propensity score, there remained no difference in SSI rates (OR 0.87, 95% confidence interval 0.36-2.06, p = 0.77).

CONCLUSION

In otherwise similar obese surgical patients weighing ≥100 kg, the administration of a pre-operative cefazolin two gram dose is associated with a similar rate of SSI compared with patients who received a three gram dose.

Impact of Intraoperative Continuous-Infusion Versus Intermittent Dosing of Cefazolin Therapy on the Incidence of Surgical Site Infections After Coronary Artery Bypass Grafting

STUDY OBJECTIVES

To determine whether intraoperative continuous-infusion (CI) cefazolin reduces the incidence of surgical site infections (SSIs) compared with intermittent (INT) cefazolin dosing in patients undergoing coronary artery bypass grafting (CABG) on cardiopulmonary bypass (CPB); safety end points and protocol adherence comparing the two dosing strategies were also explored.

DESIGN

Retrospective quasi-experimental (pre-post intervention) cohort study.

SETTING

Large academic medical center.

PATIENTS

A total of 516 adults who underwent CABG on CPB and received cefazolin intraoperatively between June 1, 2013, and December 31, 2014, were included. The INT cohort included 284 patients who underwent CABG from June 2013 to February 2014. The CI cohort included 232 patients who underwent CABG from April to December 2014, after an intraoperative CI cefazolin protocol for cardiac surgery patients undergoing CPB was adopted in March 2014.

MEASUREMENTS AND MAIN RESULTS

The primary end point was incidence of SSIs, and safety end points of renal dysfunction and seizures were evaluated. Multivariable logistic regression analysis was used to determine the impact on SSIs when controlling for other risk factors. A subgroup analysis for this study included 2 months within each time period to evaluate protocol adherence. The overall incidence of SSIs was decreased in patients receiving CI cefazolin, although this did not reach statistical significance (4.6% in the INT cohort vs 1.7% in the CI cohort, p=0.116). Superficial SSIs were significantly reduced in the CI cohort (2.8% in the INT cohort vs 0.4% in the CI cohort, p=0.039). In the regression analysis, CI cefazolin decreased the odds of SSI by 66%, although it did not reach statistical significance (p=0.077). Safety end points were not significantly different between groups. Overall protocol adherence did not differ significantly between the cohorts: 77% in the INT cohort and 67% in the CI cohort (p=0.212).

CONCLUSION

CI cefazolin significantly decreased the incidence of superficial SSIs compared with INT cefazolin in patients undergoing CABG on CPB, without increasing the risk for adverse effects. As this study was underpowered to detect a significant difference in overall SSIs, larger, randomized studies are required to validate the superiority of CI cefazolin.

Continuous Intraoperative Cefazolin Infusion May Reduce Surgical Site Infections During Cardiac Surgical Procedures: A Propensity-Matched Analysis

OBJECTIVES

The authors sought to determine whether an institutional transition from intermittent to continuous dosing of intraoperative antibiotics in cardiac surgery affected surgical site infection (SSI) outcomes.

DESIGN

A retrospective chart review utilizing propensity matching.

SETTING

A single academic, tertiary care hospital.

PARTICIPANTS

One thousand one hundred seventy-nine patients undergoing coronary artery bypass grafting (CABG) and/or cardiac valvular surgery between April 2013 and November 2014 who received perioperative cefazolin.

INTERVENTIONS

By method of cefazolin administration, patients were divided into an "intermittent-dosing" (ID) group and a "continuous-infusion" (CI) group.

MEASUREMENTS AND MAIN RESULTS

Of the 1,179 patients who underwent cardiac surgery during the study period, 1:1 propensity score matching yielded 399 patients in each group. Rates of diabetes (33.6% ID v 33.8% CI, p = 0.94), coronary artery bypass (62.3% v 61.4%, p = 0.66), and bilateral internal mammary artery harvesting (6.0% v 8.3%, p = 0.22) were similar between groups. SSIs occurred in more ID patients than CI patients (2.3% v 0.5%, p = 0.03). This difference was driven by decreases in extremity and conduit harvest site SSIs (1.8% v 0.3%, p = 0.03), as there were no episodes of mediastinitis, and superficial sternal SSI rates did not differ (0.5% v 0.3%, p = 0.56). There also were significantly fewer episodes of pneumonia in the CI group (6.0% v 2.3%, p = 0.008). Intensive care unit and total lengths of stay did not differ. Thirty-day mortality was 2.8% in both groups (p = 1.00).

CONCLUSIONS

As compared to ID regimens, CI cefazolin infusion may reduce post-cardiac surgery infectious complications. Further study in larger patient populations is needed.

Continuous versus intermittent infusions of antibiotics for the treatment of severe acute infections

BACKGROUND

Intravenous broad-spectrum antibiotics are indicated for the treatment of severe infections. However, the emergence of infections caused by multi-drug resistant organisms in conjunction with a lack of novel antibiotics has prompted the investigation of alternative dosing strategies to improve clinical efficacy and tolerability. To optimise pharmacokinetic and pharmacodynamic antibiotic parameters, continuous antibiotic infusions have been compared to traditional intermittent antibiotic infusions.

OBJECTIVES

To compare the clinical efficacy and safety of continuous intravenous administration of concentration-dependent and time-dependent antibiotics to traditional intermittent intravenous administration in adults with severe acute bacterial infections.

SEARCH METHODS

The following electronic databases were searched in September 2012: The Cochrane Injuries Group Specialised Register, Cochrane Central Register of Controlled Trials (The Cochrane Library), MEDLINE (OvidSP), EMBASE (OvidSP), CINAHL, ISI Web of Science: Science Citation Index Expanded (SCI-EXPANDED), ISI Web of Science: Conference Proceedings Citation Index-Science (CPCI-S). The reference lists of all relevant material, the Internet and the trials registry www.clinicaltrials.gov for completed and ongoing trials were also searched.

SELECTION CRITERIA

Randomized controlled trials in adults with a bacterial infection requiring intravenous antibiotic therapy comparing continuous versus intermittent infusions of antibiotics were included. Both time-dependent and concentration-dependent antibiotics were considered.

DATA COLLECTION AND ANALYSIS

Three independent authors performed data extraction for the included studies. All data was cross-checked and disagreements resolved by consensus. An intention to treat analysis was conducted using a random-effects model.

MAIN RESULTS

Twenty-nine studies met inclusion criteria with a combined total of over 1,600 patients. The majority of included studies were judged to be at unclear or high risk of bias with regard to randomisation sequence generation, allocation concealment, blinding, management of incomplete outcome data, selective outcome reporting, and other potential threats to validity. No studies were judged to be at low risk of bias for all methodological quality items assessed. There were no differences in all-cause mortality (n=1241, RR 0.89, 95% CI 0.67 - 1.20, p=0.45), infection recurrence (n=398, RR 1.22, 95% CI 0.35 - 4.19, p=0.76), clinical cure (n=975, RR 1.00, 95% CI 0.93 - 1.08, p=0.98), and superinfection post-therapy (n=813, RR 1.08, 95% CI 0.60 - 1.94, p=0.79). There were no differences in safety outcomes including adverse events (n=575, RR 1.02, 95% CI 0.94 - 1.12, p=0.63), serious adverse events (n=871, RR 1.36, 95% CI 0.80 - 2.30, p=0.26), and withdrawal due to adverse events (n=871, RR 2.03, 95% CI 0.52 - 7.95, p=0.31). A difference was observed in the subgroup analyses of clinical cure in septic versus non-septic patients, where intermittent antibiotic infusions were favoured for clinical cure in septic patients. However, this effect was not consistent between random-effects and fixed-effects analyses. No differences were found in sensitivity analyses conducted.

AUTHORS' CONCLUSIONS

There were no differences in mortality, infection recurrence, clinical cure, superinfection post-therapy, and safety outcomes when comparing continuous infusions of intravenous antibiotics to traditional intermittent infusions of antibiotics. However, the wide confidence intervals suggest that beneficial or harmful effects cannot be ruled out for all outcomes. Therefore, the current evidence is insufficient to recommend the widespread adoption of continuous infusion antibiotics in the place of intermittent infusions of antibiotics. Further large prospective randomised trials, with consistent and complete reporting of clinical outcome measures, conducted with concurrent pharmacokinetic and pharmacodynamic studies in special populations are required to determine whether adoption of continuous antibiotic infusions is warranted in specific circumstances.

Cefazolin bolus and continuous administration for elective cardiac surgery: improved pharmacokinetic and pharmacodynamic parameters

OBJECTIVE

Cefazolin (1-2 g bolus at induction possibly repeated after cardiopulmonary bypass) remains the standard for antibiotic prophylaxis in cardiac surgery. Data indicate, however, that it is underdosed with this dosing schedule. A prospective, randomized study comparing intermittent versus loading dose plus continuous infusion for the same total dose of cefazolin was performed to assess which modality is pharmacokinetically and pharmacodynamically advantageous.

METHODS

Patients received 2 g cefazolin as a starting dose and then were divided into an intermittent group (receiving another 1 g at 3, 9, and 15 hours after the first dose) and a continuous group (continuous infusion started after the first dose, providing 1 g every 6 hours for 18 hours). Cefazolin levels were measured in blood and atria.

RESULTS

Mean total and calculated free trough concentrations in blood varied greatly among patients in the intermittent group and were lower than those in the continuous group (P < .05 at 15, 18 and 24 hours). For 9 of 10 (90%) patients in the continuous infusion group, the targeted pharmacokinetic and pharmacodynamic goal (time above minimal inhibitory concentration >90%) was achieved, whereas the goal was met for only 3 of 10 (30%) in the intermittent group (P < .05). The mean atrial tissue concentration was also higher with continuous infusion (P < .05).

CONCLUSIONS

Administration of cefazolin as bolus plus continuous infusion has pharmacokinetic and pharmacodynamic advantages relative to intermittent administration. It provides more stable serum levels, lower interpatient variability, and higher myocardial tissue penetration.

Continuous cefazolin infusion to treat bone and joint infections: clinical efficacy, feasibility, safety, and serum and bone concentrations

Cefazolin has been used for many years to treat bone and joint infections. Because of its time-dependent antimicrobial activity, continuous infusion would potentially be beneficial. We report on the feasibility, safety, and efficacy of prolonged continuous intravenous cefazolin therapy in a cohort of 100 patients, their serum cefazolin levels, and the concomitant bone cefazolin concentrations in 8 of them. This retrospective cohort study included all the patients treated for bone or joint infection with a continuous cefazolin infusion administered over a 12-h period twice daily for >or=2 weeks. Drug monitoring was performed at least twice for all the patients. Serum and bone cefazolin concentrations were determined by standardized disk diffusion microbiological assays. The absence of clinical, biological, and radiological signs of infection after 2 years of follow-up and the same criteria after 1 year of follow-up defined cures and probable cures, respectively. The median treatment duration was 42 days, and the median daily cefazolin dose was 6 g. Half of the patients received parenteral antibiotic therapy on an outpatient basis. Two moderate-grade adverse events were observed. The median serum cefazolin concentrations were 63 microg/ml (range, 13 to 203 microg/ml) and 57 microg/ml (range, 29 to 128 microg/ml) on days 2 to 10 and days 11 to 21, respectively. The median bone cefazolin concentration reached 13.5 microg/g (range, 3.5 to 29 microg/g). The median bone concentration/serum concentration ratio was 0.25 (range, 0.06 to 0.41). Among 88 patients with a median follow-up of 25 months (range, 12 to 53 months), 52 were considered cured and 29 were considered probably cured. Thus, the treatment of bone and joint infections with a prolonged continuous intravenous cefazolin infusion was feasible, effective, well-tolerated, safe, and convenient, making it a strong candidate for home therapy.

Characteristics of infections in patients undergoing staged implantation for sacral nerve stimulation

OBJECTIVES

To review clinical and surgical factors in patients who have undergone staged sacral nerve stimulator implantation and to determine whether there are any identifiable risk factors for infection.

METHODS

A retrospective chart review was performed on 76 consecutive patients undergoing staged implantation for sacral nerve stimulation for voiding dysfunction. Patients with postprocedural wound infections (after Stage 1 or Stage 2) were compared with those without infections with regard to demographic factors and surgical characteristics, such as operative time and duration of exposed lead wire. Organisms cultured were also documented.

RESULTS

Lead infection occurred in 9 of 76 patients (12%). All cultures grew Staphylococcus aureus. Of 9 patients with lead infection, 6 had organisms sensitive to their perioperative antibiotic. Forty-five patients had an implantable pulse generator implanted, and 5 infections occurred (11%). Four cultures grew S. aureus (all sensitive to the perioperative antibiotic given), whereas one grew Pseudomonas. The only significant difference in clinical/surgical characteristics between infected and noninfected patients was a longer operative time for Stage 2 in infected patients. In addition, 3 patients with infection had one or more known risk factors for wound infection (steroid use, severe psoriasis, recurrent skin abscess).

CONCLUSIONS

Apart from known risk factors for surgical wound infections, the only variable we could identify that might increase the risk for infection is a longer operative time for Stage 2. S. aureus was the organism most commonly cultured. Often it was sensitive to the perioperative antibiotic prophylaxis.

Bacterial adherence to surgical sutures: can antibacterial-coated sutures reduce the risk of microbial contamination?

BACKGROUND

Surgical site infections are associated with severe morbidity and mortality. The role of surgical sutures in the etiology of surgical site infection has been the objective of discussion for decades. This study used a standardized in vitro microbiologic model to assess bacterial adherence and the antibacterial activity of a triclosan-coated polyglactin 910 (braided) suture against selected Gram-positive and Gram-negative clinical isolates that may infect surgical wounds.

STUDY DESIGN

Standardized cultures (2.0 log(10) colony forming units/mL and 5.0 log(10) colony forming units/mL of three clinical strains, Staphyllococcus aureus (methicillin-resistant S aureus [MRSA]), S epidermidis (biofilm-positive) and Escherichia coli (extended-spectrum beta-lactamase [ESBL]-producer) were inoculated to triclosan-coated and noncoated polyglactin 910 sutures to evaluate comparative adherence of bacterial isolates to the antibacterial coated and noncoated surgical sutures; to assess the impact of serum proteins (bovine serum albumin) on antibacterial activity of triclosan-coated suture; and to document the duration of antibacterial activity of the triclosan-coated material. Selected suture samples were prepared for scanning electron microscopy to demonstrate bacterial adherence.

RESULTS

Substantial (p < 0.01) reductions in both Gram-positive and Gram-negative bacterial adherence were observed on triclosan-coated sutures compared with noncoated material. Pretreatment of surgical sutures with 20% BSA did not diminish antibacterial activity of the triclosan-coated braided device compared with noncoated suture (p < 0.01), and antibacterial activity was documented to persist for at least 96 hours compared with controls (p < 0.01).

CONCLUSIONS

The in vitro model demonstrated a considerable reduction (p < 0.01) in Gram-positive and Gram-negative bacterial adherence to a triclosan-coated braided suture, which was associated with decreased microbial viability (p < 0.001). Because bacterial contamination of suture material within a surgical wound may increase the virulence of a surgical site infection, treating the suture with triclosan provides an effective strategy for reducing perioperative surgical morbidity.

Continuous versus intermittent intravenous administration of antibacterials with time-dependent action: a systematic review of pharmacokinetic and pharmacodynamic parameters

We performed a systematic review of randomised clinical trials to evaluate the comparative pharmacokinetic and pharmacodynamic properties of the continuous versus intermittent mode of intravenous administration of various antibacterials. Data were identified from PubMed (January 1950 to January 2005), Current Contents, the Cochrane central register of controlled trials, and references from relevant articles and reviews. Seventeen randomised clinical trials comparing continuous with intermittent intravenous administration of the same antibacterial regimen and examining the pharmacokinetic and pharmacodynamic properties were included in this systematic review. We reviewed data regarding the clinical setting, number of participants, antibacterial agents and dosages used, as well as maximum serum concentration (Cmax), trough serum concentration (Cmin), steady-state or plateau serum concentration (Css), area under the concentration-time curve (AUC), time above the minimum inhibitory concentration (MIC) [T>MIC], AUC: MIC, elimination rate constant, elimination half-life, volume of distribution and systematic clearance. The mean Cmax of the intermittently administered antibacterials was higher compared with Css achieved by the continuous infusion of the same antibacterial in all eligible studies (Cmax was on average 5.5 times higher than Css, range 1.9-11.2). Css was on average 5.8 times higher than the Cmin of the intermittently administered antibacterials (range 1.2-15.6). In three of six studies the length of time that the drug concentration was above the MIC of the responsible pathogens was longer in patients receiving the antibacterials continuously. In conclusion, the reviewed data suggest that the continuous intravenous infusion of antibacterials with time-dependent bacterial killing seems to be superior than the intermittent intravenous administration, from a pharmacodynamic point of view, at least when treating bacteria with high MIC values for the studied antibacterials.

Continuous versus intermittent intravenous administration of antibiotics: a meta-analysis of randomised controlled trials

Intermittent intravenous administration of antibiotics is the first-line approach in the management of severe infections worldwide. However, the potential benefits of alternate modes of administration of antibiotics, including continuous intravenous infusion, deserve further evaluation. We did a meta-analysis of randomised controlled trials comparing continuous intravenous infusion with intermittent intravenous administration of the same antibiotic regimen. Nine randomised controlled trials studying beta-lactams, aminoglycosides, and vancomycin were included. Clinical failure was lower, although without statistical significance, in patients receiving continuous infusion of antibiotics (pooled OR 0.73, 95% CI 0.53-1.01); the difference was statistically significant in a subset of randomised controlled trials that used the same total daily antibiotic dose for both intervention arms (0.70, 0.50-0.98, fixed and random effects models). Regarding mortality and nephrotoxicity, no differences were found (mortality 0.89, 0.48-1.64; nephrotoxicity 0.91, 0.56-1.47). In conclusion, the data suggest that the administration of the same total antibiotic dose by continuous intravenous infusion may be more efficient, with regard to clinical effectiveness, compared with the intermittent mode. In an era of gradually increasing resistance among most pathogens, the potential advantages of continuous intravenous administration of antibiotics on several clinical outcomes should be further investigated.