- Email: [email protected]
UTILIZATION AND OUTCOMES OF MINIMALLY INVASIVE RADICAL PROSTATECTOMY
4
THE JOURNAL OF UROLOGY®
RESULTS: There were 1237 men who underwent RP from 1994 - 2007 and had complete information. Overall, 74.5% of men XQGHUZHQW/1'ZLWK53/1'XWLOL]DWLRQZDVQRWGLIIHUHQWE\SDWLHQWDJH (p = 0.401). The proportion of men undergoing LND decreased from 81% in 1994 - 1997 to 70.0% in 2004 - 2007 (p = 0.001). Center was VLJQL¿FDQWO\DVVRFLDWHGZLWKSHUIRUPDQFHRI/1'ZLWKWKHSURSRUWLRQRI RP accompanied by LND ranging from 40.2% to 100% (p < 0.001). Sixty¿YHSHUFHQWRIORZULVNPHQXQGHUZHQW/1'YHUVXVRILQWHUPHGLDWH risk and 92% of high risk men (p<0.001). On multivariable analysis, only FHQWHUDQGULVNJURXSZHUHVLJQL¿FDQWO\DVVRFLDWHGZLWKSHUIRUPDQFHRI LND with RP. Intermediate and high risk men were more likely to undergo LND than low risk men (Table, p<0.001). The results of this model were similar when the center with a 100% LND rate was excluded. Multivariate Odds Ratios for Undergoing LND by Risk Group &RQ¿GHQFH Risk Group Odds Ratio Interval Low 1.00 (referent) Intermediate 2.80 1.95 - 4.02 High 11.1 6.14 - 20.1
&21&/86,216 7KHSUDFWLFHRI/1'DW53UHÀHFWVFXUUHQW XQGHUVWDQGLQJRISURVWDWHFDQFHUULVNVWUDWL¿FDWLRQGHVSLWHFRQWURYHUV\ regarding the impact of LND on biochemical recurrence. However, VLJQL¿FDQW ORFDO YDULDWLRQ LQ SUDFWLFH UHPDLQV )XUWKHU LQYHVWLJDWLRQ LV UHTXLUHG WR GH¿QH WKH LPSDFW RI /1' RQ ELRFKHPLFDO IDLOXUH DQG WR understand patient, physician, and local market factors associated with /1'XWLOL]DWLRQ Source of Funding: None
9 FACTORS ASSOCIATED WITH CHANGES IN BENIGN PROSTATIC HYPERPLASIA (BPH) MANAGEMENT IN CLINICAL PRACTICE: LONGITUDINAL RESULTS FROM THE BPH REGISTRY Martin Miner*, John T Wei, Claus G Roehrborn, Allen D Seftel, Raymond C Rosen. Swansea, MA, Ann Arbor, MI, Dallas, TX, Cleveland, OH, and Watertown, MA. INTRODUCTION AND OBJECTIVE: Medical therapy (MT) is a ¿UVWOLQHPDQDJHPHQWDSSURDFKIRUERWKHUVRPHPDOHORZHUXULQDU\WUDFW symptoms (LUTS). Treatment patterns for LUTS/BPH in clinical practice DUH QRW ZHOO FKDUDFWHUL]HG7KH %3+ 5HJLVWU\ LV WKH ¿UVW PXOWLFHQWHU observational, disease registry to collect treatment and outcomes data on 6909 men with LUTS/BPH managed by US urologists (UROs) and primary care physicians (PCPs) (Roehrborn et al. BJU Int 2007). We used BPH Registry data to identify determinants of changes in BPH management for men on MT or watchful waiting (WW). METHODS: The BPH Registry is a prospective, observational, disease registry collecting demographic, clinical, and quality-of-life data and tracking management practices by UROs and PCPs. This analysis LQFOXGHGPHQZLWKIROORZXSYLVLWZKRZHUHIROORZHGWRWKHLUODVWYLVLW within 400 days of baseline or to a BPH management change, whichever FDPH¿UVW$%3+PDQDJHPHQWFKDQJHZDVGH¿QHGDVDFKDQJHIURP WW to MT or from 1 MT category to another (selective alpha-blocker >DOIX]RVLQRUWDPVXORVLQ@QRQVHOHFWLYHDOSKDEORFNHUDOSKDUHGXFWDVH LQKLELWRU>$5,@RUDOSKDEORFNHU$5,FRPELQDWLRQWKHUDS\ 0HQRQ anticholinergics were excluded. Bivariate analyses and multivariate Cox SURSRUWLRQDOKD]DUGVPRGHOVZHUHXVHGWRLGHQWLI\IDFWRUVLQGHSHQGHQWO\ predicting a change in BPH management. RESULTS: Of 3764 men (mean follow-up: 254 days), 379 (10%) had a change in BPH management. In multivariate adjusted models, greater LUTS and LUTS bother, a higher Sexual Health Inventory for Men (SHIM) score, and management by a URO were associated with a change from WW to MT, whereas higher income, a higher BPH Impact Index, and a higher SHIM score were associated with a change from MT to another MT (Table). CONCLUSIONS: In the BPH Registry “real-world” population, greater LUTS severity and bother, management by a URO, worse quality of life, better sexual function, and higher income were associated with changes in BPH management.
Vol. 179, No. 4, Supplement, Saturday, May 17, 2008
6LJQL¿FDQW3UHGLFWRUVRID&KDQJHLQ%3+0DQDJHPHQW 0XOWLYDULDWH Analyses) WW ĺ MT MT ĺ Another MT Hazard Ratio (95% CI) Physician type (URO vs PCP) 2.6 (1.2-5.7) NS ,QFRPH YV NS 1.8 (1.0-3.3) IPSS (higher vs lower)* 1.1 (1.0-1.1) NS IPSS bother score (higher vs lower)* 1.5 (1.1-2.1) NS BPH Impact Index (higher vs lower)* NS 1.2 (1.2-1.3) SHIM score (higher vs lower)† 1.1 (1.0-1.2) 1.1 (1.0-1.1)
+LJKHUVFRUHLQGLFDWHVZRUVH/876ERWKHURUTXDOLW\RIOLIH +LJKHU VFRUHLQGLFDWHVEHWWHUHUHFWLOHIXQFWLRQ 16 QRWVLJQL¿FDQW
Source of Funding:VDQR¿DYHQWLV
10 ACCURACY OF CLAIMS DATA TO IDENTIFY INCIDENT CASES OF BENIGN PROSTATIC HYPERPLASIA Simon P Kim*, John M Hollingsworth, Brent K Hollenbeck, David A Hanauer, Kellie S Paich, John T Wei. Ann Arbor, MI. INTRODUCTION AND OBJECTIVE: Claims data are increasingly used to measure the quality of patient care. In the context of benign prostatic hyperplasia (BPH) , identifying incident cases would be particularly important, especially for attribution of care. For this reason, we used explicit medical record review to measure the accuracy of claims-based algorithms to ascertain incident cases of BPH. METHODS: Patients with at least 1 ICD-9 diagnosis code for BPH in a managed care setting between 1997 and 2005. Using diagnosis code-based algorithms, incident cases of patients with BPH were LGHQWL¿HG$QH[SOLFLWFKDUWUHYLHZZDVSHUIRUPHGRQDUDQGRPVDPSOH of 96 claims and used as the gold standard. Clinical data included the diagnosis date, symptoms, diagnostic tests, primary treating physician specialty and the treatment course. Agreement between claims and clinical data was assessed. Among those with BPH documentation in the medical record, we examined whether or not age, specialty and treatments were associated with an accurate administrative claims. RESULTS: Of the 96 incident cases of BPH ascertained in FODLPV ZH GHWHUPLQHG WKDW SDWLHQWV ZHUH PLVFODVVL¿HG including 26 (27.1%) patients without any supporting clinical data. Further, 20 patients (20.8%) proved to be prevalent cases. Comparisons EHWZHHQLQFLGHQWDQGSUHYDOHQWFDVHVUHYHDOHGWKDWPLVFODVVL¿FDWLRQZDV more likely among older patients but did not vary according to specialty or treatment course. CONCLUSIONS: Our initial claims-based algorithm to LGHQWLI\LQFLGHQWFDVHVRI%3+PLVFODVVL¿HG RISDWLHQWV)XUWKHU UH¿QHPHQWV LQ RXU DOJRULWKP ZLWK PRUH VWULQJHQW LQFOXVLRQ FULWHULD will likely result in a population of incident BPH cases with a lower PLVFODVVL¿FDWLRQUDWH8OWLPDWHO\WKHVHGDWDZLOOEHXVHIXOIRUVWXG\LQJ quality of BPH care in administrative data. Source of Funding: None
11 UTILIZATION AND OUTCOMES OF MINIMALLY INVASIVE RADICAL PROSTATECTOMY Michaella M Prasad*, Jim C Hu, Qin Wang, Chris L Pashos, Stuart R Lipsitz, Nancy L Keating. Boston, MA, and Bethesda, MD. INTRODUCTION AND OBJECTIVE: Demand for minimally invasive radical prostatectomy (MIRP) to treat prostate cancer is LQFUHDVLQJ KRZHYHURXWFRPHVUHPDLQXQFOHDU:HDVVHVVHGXWLOL]DWLRQ complications, lengths of stay, and salvage therapy rates for minimallyinvasive vs. open radical prostatectomy and assessed whether MIRP surgeon volume is associated with better outcomes. 0(7+2'6 :HLGHQWL¿HGPHQXQGHUJRLQJ0,53DQG open radical prostatectomy during 2003-05 from a national 5% sample RI 0HGLFDUH EHQH¿FLDULHV :H DVVHVVHG WKH DVVRFLDWLRQ EHWZHHQ surgical approach and outcomes, adjusting for surgeon volume, age, race, comorbidity, and geographic region. 5(68/76 0,53XWLOL]DWLRQLQFUHDVHGIURPLQWR 31.4% in 2005. Men undergoing MIRP vs. open radical prostatectomy had fewer perioperative complications (29.8% vs. 36.4%, p=0.002) and VKRUWHUOHQJWKVRIVWD\ YVGD\V S KRZHYHUWKH\ZHUH more likely to receive salvage therapy (27.8% vs. 9.1%, p<0.001). In adjusted analyses, MIRP vs. open radical prostatectomy was associated
Vol. 179, No. 4, Supplement, Saturday, May 17, 2008
ZLWK IHZHU SHULRSHUDWLYH FRPSOLFDWLRQV RGGV UDWLR >25@ FRQ¿GHQFHLQWHUYDO>&,@ VKRUWHUOHQJWKVRIVWD\ SDUDPHWHU estimate -2.99, 95%CI -3.45,-2.53) but more anastomotic strictures (OR 1.40, 95%CI 1.04,1.87) and higher rates of salvage therapy (OR 3.67, 95%CI 2.81,4.81). Patients of high-volume MIRP experienced fewer anastomotic strictures (OR 0.93, 95%CI 0.87,0.99) and less salvage therapy (OR 0.92, 95%CI 0.88,0.98). CONCLUSIONS: Men undergoing MIRP vs. open radical prostatectomy have lower risk for perioperative complications and shorter lengths of stay, but are at higher risk for salvage therapy and anastomotic strictures. However, risk for these unfavorable outcomes decreases with increasing MIRP surgical volume.
THE JOURNAL OF UROLOGY®
Total lifetime number of cases performed < 50 50-99 100-249 Total
5
Number of surgeons (%) Excluding prior opens 16 (39) 4 (10) 11 (27) 10 (24) 41 (100)
Including prior opens 13 (32) 6 (15) 9 (22) 13 (32) 41 (100)
Positive surgical margins were reported in 1291 (21%) patients. The surgical learning curve for surgical margins with laparoscopic RP LV VKRZQ LQ WKH ¿JXUH EHORZ 7KH [D[LV H[FOXGHV SULRU RSHQ FDVHV performed. On multivariable analysis, higher surgeon experience was VLJQL¿FDQWO\ DVVRFLDWHG ZLWK D ORZHU ULVN RI SRVLWLYH VXUJLFDO PDUJLQV S 7KHQXPEHURISULRURSHQ53VZDVQRWDVWDWLVWLFDOO\VLJQL¿FDQW predictor in the multivariable model. CONCLUSIONS: The learning curve for surgical margins after LRP continues to improve even with a large number of prior surgeries. Prior open experience does not improve margin rates suggesting that WKHVHDUHSULPDULO\DIXQFWLRQRIVSHFL¿FDOO\ODSDURVFRSLFH[SHULHQFH Source of Funding: None
13
Source of Funding: A Lance Armstrong Young Investigator Award granted to Dr. Hu was used to fund this study.
12 LEARNING CURVE OF POSITIVE MARGIN RATE IN LAPAROSCOPIC RADICAL PROSTATECTOMY Fernando P Secin, Angel M Cronin, Jens Rassweiler, M Hruza, Clement-Claude C Abbou, Alexandre de la Taille, Laurent Salomon, Gunter Janetschek, Faisal Nassar, Ingolf Turk, Alex J Vanni, Inderbir S Gill, Philippe Koenig, Jihad H Kaouk, Luis Martinez Pineiro, Vito Pansadoro, Paolo Emiliozzi, Anders Bjartell, Christopher G Eden, A Richards, Roland Van Velthoven, Caroline J Savage, Andrew J Vickers, Karim Touijer, Bertrand D Guillonneau*. New York, NY, Heidelberg, Germany, Paris, France, Linz, Austria, Burlington, MA, Cleveland, OH, Madrid, Spain, Rome, Italy, Malmo, Sweden, London, United Kingdom, and Brussels, Belgium. INTRODUCTION AND OBJECTIVE: The learning curve of laparoscopic radical prostatectomy (LRP) for positive margins has QRWEHHQZHOOGH¿QHG,QWKLVVWXG\ZHVRXJKWWRGHWHUPLQHWKHVKDSH RIWKHOHDUQLQJFXUYHDQGH[DPLQHZKHWKHUWKLVLVLQÀXHQFHGE\SULRU experience with open RP. METHODS: The data base includes 9237 prostate cancer patients treated by 53 surgeons from 14 institutions in both Europe and the USA. All patients treated laparoscopically by a participating surgeon were included. Surgeons gave an estimate of the number of RSHQ53¶VWKH\FRQGXFWHGEHIRUHWKH¿UVW/53$IWHUH[FOXVLRQRI patients for missing data, the study cohort consisted of 6274 patients. The probability of a positive margin was calculated as a function of VXUJHRQH[SHULHQFHGH¿QHGDVWKHQXPEHURI/53FRQGXFWHGE\WKH surgeon prior to the index case - with adjustment for stage, grade, PSA and year of surgery. RESULTS: The distribution of surgeons by the total lifetime number of LRP is shown in the table below. The most experienced surgeon completed 1434 cases with and 1066 cases without inclusion of prior open surgeries.
RECOVERY AFTER ABDOMINAL AND PELVIC SURGERY Brent K Hollenbeck*, John M Hollingsworth, Rodney L Dunn, J Stuart Wolf, James E Montie, Scott M Gilbert, Alon Z Weizer, Martin G Sanda, John T Wei. Ann Arbor, MI, and Boston, MA. INTRODUCTION AND OBJECTIVE: Some surgical LQQRYDWLRQV DUH WRXWHG WR LPSURYH UHFRYHU\ KRZHYHU WKHLU DGRSWLRQ may be associated with a learning curve and unforeseen consequences. We used a validated instrument, the Convalescence And Recovery Evaluation (CARE), to measure return to baseline after surgery and explored clinical factors associated with faster recovery. METHODS: Using CARE, patient health was measured at baseline and at 5 intervals postoperatively among 96 patients undergoing abdominal and pelvic surgery (response rate=90%). CARE PHDVXUHVUHFRYHU\LQGLPHQVLRQVDQGLVVXPPDUL]HGDVDFRPSRVLWH score (with higher scores representing a better health state). Patients were sorted into groups depending on when they achieved recovery to 90% of baseline (< 2 weeks, 2 to 4 weeks, > 4 weeks). Chi square tests and logistic models were used to measure relationships between time to recovery and patient characteristics, processes of care and outcomes. RESULTS: 44%, 28%, and 28% of patients reached 90% of their baseline health status within 2 weeks, between 2 to 4 weeks, and at > 4 weeks, respectively. As illustrated in the Table, patients who recovered faster were generally younger, male, single and undergoing ambulatory laparoscopic surgery (all p<0.05). Patients who were married (adjusted OR 3.2, 95% CI 1.1-9.8) and those who underwent surgery for cancer (adjusted OR 3.9, 95% CI 1.2-12.2) were more likely to require > 4 weeks to recover. CONCLUSIONS: In this heterogeneous population, we observed many clinical factors associated with time to recovery to baseline health after abdominal and pelvic surgery. CARE appears to be useful for discriminating rates of recovery and therefore may be sensitive to the consequences of the learning curve for surgical innovations.
THE JOURNAL OF UROLOGY®
RESULTS: There were 1237 men who underwent RP from 1994 - 2007 and had complete information. Overall, 74.5% of men XQGHUZHQW/1'ZLWK53/1'XWLOL]DWLRQZDVQRWGLIIHUHQWE\SDWLHQWDJH (p = 0.401). The proportion of men undergoing LND decreased from 81% in 1994 - 1997 to 70.0% in 2004 - 2007 (p = 0.001). Center was VLJQL¿FDQWO\DVVRFLDWHGZLWKSHUIRUPDQFHRI/1'ZLWKWKHSURSRUWLRQRI RP accompanied by LND ranging from 40.2% to 100% (p < 0.001). Sixty¿YHSHUFHQWRIORZULVNPHQXQGHUZHQW/1'YHUVXVRILQWHUPHGLDWH risk and 92% of high risk men (p<0.001). On multivariable analysis, only FHQWHUDQGULVNJURXSZHUHVLJQL¿FDQWO\DVVRFLDWHGZLWKSHUIRUPDQFHRI LND with RP. Intermediate and high risk men were more likely to undergo LND than low risk men (Table, p<0.001). The results of this model were similar when the center with a 100% LND rate was excluded. Multivariate Odds Ratios for Undergoing LND by Risk Group &RQ¿GHQFH Risk Group Odds Ratio Interval Low 1.00 (referent) Intermediate 2.80 1.95 - 4.02 High 11.1 6.14 - 20.1
&21&/86,216 7KHSUDFWLFHRI/1'DW53UHÀHFWVFXUUHQW XQGHUVWDQGLQJRISURVWDWHFDQFHUULVNVWUDWL¿FDWLRQGHVSLWHFRQWURYHUV\ regarding the impact of LND on biochemical recurrence. However, VLJQL¿FDQW ORFDO YDULDWLRQ LQ SUDFWLFH UHPDLQV )XUWKHU LQYHVWLJDWLRQ LV UHTXLUHG WR GH¿QH WKH LPSDFW RI /1' RQ ELRFKHPLFDO IDLOXUH DQG WR understand patient, physician, and local market factors associated with /1'XWLOL]DWLRQ Source of Funding: None
9 FACTORS ASSOCIATED WITH CHANGES IN BENIGN PROSTATIC HYPERPLASIA (BPH) MANAGEMENT IN CLINICAL PRACTICE: LONGITUDINAL RESULTS FROM THE BPH REGISTRY Martin Miner*, John T Wei, Claus G Roehrborn, Allen D Seftel, Raymond C Rosen. Swansea, MA, Ann Arbor, MI, Dallas, TX, Cleveland, OH, and Watertown, MA. INTRODUCTION AND OBJECTIVE: Medical therapy (MT) is a ¿UVWOLQHPDQDJHPHQWDSSURDFKIRUERWKHUVRPHPDOHORZHUXULQDU\WUDFW symptoms (LUTS). Treatment patterns for LUTS/BPH in clinical practice DUH QRW ZHOO FKDUDFWHUL]HG7KH %3+ 5HJLVWU\ LV WKH ¿UVW PXOWLFHQWHU observational, disease registry to collect treatment and outcomes data on 6909 men with LUTS/BPH managed by US urologists (UROs) and primary care physicians (PCPs) (Roehrborn et al. BJU Int 2007). We used BPH Registry data to identify determinants of changes in BPH management for men on MT or watchful waiting (WW). METHODS: The BPH Registry is a prospective, observational, disease registry collecting demographic, clinical, and quality-of-life data and tracking management practices by UROs and PCPs. This analysis LQFOXGHGPHQZLWKIROORZXSYLVLWZKRZHUHIROORZHGWRWKHLUODVWYLVLW within 400 days of baseline or to a BPH management change, whichever FDPH¿UVW$%3+PDQDJHPHQWFKDQJHZDVGH¿QHGDVDFKDQJHIURP WW to MT or from 1 MT category to another (selective alpha-blocker >DOIX]RVLQRUWDPVXORVLQ@QRQVHOHFWLYHDOSKDEORFNHUDOSKDUHGXFWDVH LQKLELWRU>$5,@RUDOSKDEORFNHU$5,FRPELQDWLRQWKHUDS\ 0HQRQ anticholinergics were excluded. Bivariate analyses and multivariate Cox SURSRUWLRQDOKD]DUGVPRGHOVZHUHXVHGWRLGHQWLI\IDFWRUVLQGHSHQGHQWO\ predicting a change in BPH management. RESULTS: Of 3764 men (mean follow-up: 254 days), 379 (10%) had a change in BPH management. In multivariate adjusted models, greater LUTS and LUTS bother, a higher Sexual Health Inventory for Men (SHIM) score, and management by a URO were associated with a change from WW to MT, whereas higher income, a higher BPH Impact Index, and a higher SHIM score were associated with a change from MT to another MT (Table). CONCLUSIONS: In the BPH Registry “real-world” population, greater LUTS severity and bother, management by a URO, worse quality of life, better sexual function, and higher income were associated with changes in BPH management.
Vol. 179, No. 4, Supplement, Saturday, May 17, 2008
6LJQL¿FDQW3UHGLFWRUVRID&KDQJHLQ%3+0DQDJHPHQW 0XOWLYDULDWH Analyses) WW ĺ MT MT ĺ Another MT Hazard Ratio (95% CI) Physician type (URO vs PCP) 2.6 (1.2-5.7) NS ,QFRPH YV NS 1.8 (1.0-3.3) IPSS (higher vs lower)* 1.1 (1.0-1.1) NS IPSS bother score (higher vs lower)* 1.5 (1.1-2.1) NS BPH Impact Index (higher vs lower)* NS 1.2 (1.2-1.3) SHIM score (higher vs lower)† 1.1 (1.0-1.2) 1.1 (1.0-1.1)
+LJKHUVFRUHLQGLFDWHVZRUVH/876ERWKHURUTXDOLW\RIOLIH +LJKHU VFRUHLQGLFDWHVEHWWHUHUHFWLOHIXQFWLRQ 16 QRWVLJQL¿FDQW
Source of Funding:VDQR¿DYHQWLV
10 ACCURACY OF CLAIMS DATA TO IDENTIFY INCIDENT CASES OF BENIGN PROSTATIC HYPERPLASIA Simon P Kim*, John M Hollingsworth, Brent K Hollenbeck, David A Hanauer, Kellie S Paich, John T Wei. Ann Arbor, MI. INTRODUCTION AND OBJECTIVE: Claims data are increasingly used to measure the quality of patient care. In the context of benign prostatic hyperplasia (BPH) , identifying incident cases would be particularly important, especially for attribution of care. For this reason, we used explicit medical record review to measure the accuracy of claims-based algorithms to ascertain incident cases of BPH. METHODS: Patients with at least 1 ICD-9 diagnosis code for BPH in a managed care setting between 1997 and 2005. Using diagnosis code-based algorithms, incident cases of patients with BPH were LGHQWL¿HG$QH[SOLFLWFKDUWUHYLHZZDVSHUIRUPHGRQDUDQGRPVDPSOH of 96 claims and used as the gold standard. Clinical data included the diagnosis date, symptoms, diagnostic tests, primary treating physician specialty and the treatment course. Agreement between claims and clinical data was assessed. Among those with BPH documentation in the medical record, we examined whether or not age, specialty and treatments were associated with an accurate administrative claims. RESULTS: Of the 96 incident cases of BPH ascertained in FODLPV ZH GHWHUPLQHG WKDW SDWLHQWV ZHUH PLVFODVVL¿HG including 26 (27.1%) patients without any supporting clinical data. Further, 20 patients (20.8%) proved to be prevalent cases. Comparisons EHWZHHQLQFLGHQWDQGSUHYDOHQWFDVHVUHYHDOHGWKDWPLVFODVVL¿FDWLRQZDV more likely among older patients but did not vary according to specialty or treatment course. CONCLUSIONS: Our initial claims-based algorithm to LGHQWLI\LQFLGHQWFDVHVRI%3+PLVFODVVL¿HG RISDWLHQWV)XUWKHU UH¿QHPHQWV LQ RXU DOJRULWKP ZLWK PRUH VWULQJHQW LQFOXVLRQ FULWHULD will likely result in a population of incident BPH cases with a lower PLVFODVVL¿FDWLRQUDWH8OWLPDWHO\WKHVHGDWDZLOOEHXVHIXOIRUVWXG\LQJ quality of BPH care in administrative data. Source of Funding: None
11 UTILIZATION AND OUTCOMES OF MINIMALLY INVASIVE RADICAL PROSTATECTOMY Michaella M Prasad*, Jim C Hu, Qin Wang, Chris L Pashos, Stuart R Lipsitz, Nancy L Keating. Boston, MA, and Bethesda, MD. INTRODUCTION AND OBJECTIVE: Demand for minimally invasive radical prostatectomy (MIRP) to treat prostate cancer is LQFUHDVLQJ KRZHYHURXWFRPHVUHPDLQXQFOHDU:HDVVHVVHGXWLOL]DWLRQ complications, lengths of stay, and salvage therapy rates for minimallyinvasive vs. open radical prostatectomy and assessed whether MIRP surgeon volume is associated with better outcomes. 0(7+2'6 :HLGHQWL¿HGPHQXQGHUJRLQJ0,53DQG open radical prostatectomy during 2003-05 from a national 5% sample RI 0HGLFDUH EHQH¿FLDULHV :H DVVHVVHG WKH DVVRFLDWLRQ EHWZHHQ surgical approach and outcomes, adjusting for surgeon volume, age, race, comorbidity, and geographic region. 5(68/76 0,53XWLOL]DWLRQLQFUHDVHGIURPLQWR 31.4% in 2005. Men undergoing MIRP vs. open radical prostatectomy had fewer perioperative complications (29.8% vs. 36.4%, p=0.002) and VKRUWHUOHQJWKVRIVWD\ YVGD\V S KRZHYHUWKH\ZHUH more likely to receive salvage therapy (27.8% vs. 9.1%, p<0.001). In adjusted analyses, MIRP vs. open radical prostatectomy was associated
Vol. 179, No. 4, Supplement, Saturday, May 17, 2008
ZLWK IHZHU SHULRSHUDWLYH FRPSOLFDWLRQV RGGV UDWLR >25@ FRQ¿GHQFHLQWHUYDO>&,@ VKRUWHUOHQJWKVRIVWD\ SDUDPHWHU estimate -2.99, 95%CI -3.45,-2.53) but more anastomotic strictures (OR 1.40, 95%CI 1.04,1.87) and higher rates of salvage therapy (OR 3.67, 95%CI 2.81,4.81). Patients of high-volume MIRP experienced fewer anastomotic strictures (OR 0.93, 95%CI 0.87,0.99) and less salvage therapy (OR 0.92, 95%CI 0.88,0.98). CONCLUSIONS: Men undergoing MIRP vs. open radical prostatectomy have lower risk for perioperative complications and shorter lengths of stay, but are at higher risk for salvage therapy and anastomotic strictures. However, risk for these unfavorable outcomes decreases with increasing MIRP surgical volume.
THE JOURNAL OF UROLOGY®
Total lifetime number of cases performed < 50 50-99 100-249 Total
5
Number of surgeons (%) Excluding prior opens 16 (39) 4 (10) 11 (27) 10 (24) 41 (100)
Including prior opens 13 (32) 6 (15) 9 (22) 13 (32) 41 (100)
Positive surgical margins were reported in 1291 (21%) patients. The surgical learning curve for surgical margins with laparoscopic RP LV VKRZQ LQ WKH ¿JXUH EHORZ 7KH [D[LV H[FOXGHV SULRU RSHQ FDVHV performed. On multivariable analysis, higher surgeon experience was VLJQL¿FDQWO\ DVVRFLDWHG ZLWK D ORZHU ULVN RI SRVLWLYH VXUJLFDO PDUJLQV S 7KHQXPEHURISULRURSHQ53VZDVQRWDVWDWLVWLFDOO\VLJQL¿FDQW predictor in the multivariable model. CONCLUSIONS: The learning curve for surgical margins after LRP continues to improve even with a large number of prior surgeries. Prior open experience does not improve margin rates suggesting that WKHVHDUHSULPDULO\DIXQFWLRQRIVSHFL¿FDOO\ODSDURVFRSLFH[SHULHQFH Source of Funding: None
13
Source of Funding: A Lance Armstrong Young Investigator Award granted to Dr. Hu was used to fund this study.
12 LEARNING CURVE OF POSITIVE MARGIN RATE IN LAPAROSCOPIC RADICAL PROSTATECTOMY Fernando P Secin, Angel M Cronin, Jens Rassweiler, M Hruza, Clement-Claude C Abbou, Alexandre de la Taille, Laurent Salomon, Gunter Janetschek, Faisal Nassar, Ingolf Turk, Alex J Vanni, Inderbir S Gill, Philippe Koenig, Jihad H Kaouk, Luis Martinez Pineiro, Vito Pansadoro, Paolo Emiliozzi, Anders Bjartell, Christopher G Eden, A Richards, Roland Van Velthoven, Caroline J Savage, Andrew J Vickers, Karim Touijer, Bertrand D Guillonneau*. New York, NY, Heidelberg, Germany, Paris, France, Linz, Austria, Burlington, MA, Cleveland, OH, Madrid, Spain, Rome, Italy, Malmo, Sweden, London, United Kingdom, and Brussels, Belgium. INTRODUCTION AND OBJECTIVE: The learning curve of laparoscopic radical prostatectomy (LRP) for positive margins has QRWEHHQZHOOGH¿QHG,QWKLVVWXG\ZHVRXJKWWRGHWHUPLQHWKHVKDSH RIWKHOHDUQLQJFXUYHDQGH[DPLQHZKHWKHUWKLVLVLQÀXHQFHGE\SULRU experience with open RP. METHODS: The data base includes 9237 prostate cancer patients treated by 53 surgeons from 14 institutions in both Europe and the USA. All patients treated laparoscopically by a participating surgeon were included. Surgeons gave an estimate of the number of RSHQ53¶VWKH\FRQGXFWHGEHIRUHWKH¿UVW/53$IWHUH[FOXVLRQRI patients for missing data, the study cohort consisted of 6274 patients. The probability of a positive margin was calculated as a function of VXUJHRQH[SHULHQFHGH¿QHGDVWKHQXPEHURI/53FRQGXFWHGE\WKH surgeon prior to the index case - with adjustment for stage, grade, PSA and year of surgery. RESULTS: The distribution of surgeons by the total lifetime number of LRP is shown in the table below. The most experienced surgeon completed 1434 cases with and 1066 cases without inclusion of prior open surgeries.
RECOVERY AFTER ABDOMINAL AND PELVIC SURGERY Brent K Hollenbeck*, John M Hollingsworth, Rodney L Dunn, J Stuart Wolf, James E Montie, Scott M Gilbert, Alon Z Weizer, Martin G Sanda, John T Wei. Ann Arbor, MI, and Boston, MA. INTRODUCTION AND OBJECTIVE: Some surgical LQQRYDWLRQV DUH WRXWHG WR LPSURYH UHFRYHU\ KRZHYHU WKHLU DGRSWLRQ may be associated with a learning curve and unforeseen consequences. We used a validated instrument, the Convalescence And Recovery Evaluation (CARE), to measure return to baseline after surgery and explored clinical factors associated with faster recovery. METHODS: Using CARE, patient health was measured at baseline and at 5 intervals postoperatively among 96 patients undergoing abdominal and pelvic surgery (response rate=90%). CARE PHDVXUHVUHFRYHU\LQGLPHQVLRQVDQGLVVXPPDUL]HGDVDFRPSRVLWH score (with higher scores representing a better health state). Patients were sorted into groups depending on when they achieved recovery to 90% of baseline (< 2 weeks, 2 to 4 weeks, > 4 weeks). Chi square tests and logistic models were used to measure relationships between time to recovery and patient characteristics, processes of care and outcomes. RESULTS: 44%, 28%, and 28% of patients reached 90% of their baseline health status within 2 weeks, between 2 to 4 weeks, and at > 4 weeks, respectively. As illustrated in the Table, patients who recovered faster were generally younger, male, single and undergoing ambulatory laparoscopic surgery (all p<0.05). Patients who were married (adjusted OR 3.2, 95% CI 1.1-9.8) and those who underwent surgery for cancer (adjusted OR 3.9, 95% CI 1.2-12.2) were more likely to require > 4 weeks to recover. CONCLUSIONS: In this heterogeneous population, we observed many clinical factors associated with time to recovery to baseline health after abdominal and pelvic surgery. CARE appears to be useful for discriminating rates of recovery and therefore may be sensitive to the consequences of the learning curve for surgical innovations.