Você está na página 1de 8

@ 2OO4 American Society of Anesthesiologists, Inc.

rvilliams & Wilkiro, Inc
Anesthesiology 2004; 100:98 - I 05

Cbronic Pain Managetnent

American SocieQt of Anestltesiologists Closed Claims Proiect
Dermot R. FiEgibbon, M.D.,* Karen L. Posner, Ph.D.,I Karen B. Domino, M.D., M.P.H.,| Robert A. Caplan, M.D.,5
Lorri A. Lee, M.D.,ll Frederick W. Cheney, M.D.#

Backgroun* The practlce of chronic paln management has sia-based chronic pain medicine typically involves pa-
grown steadlly ln recent years. The purpose ofthls studywas to tient evaluation; provision and interpretation of diagnos-
tdentlfy and descrlbe lssues and trends ln liabiltty related to tic procedures; clinical phamacology; and provision of
chronlc paln management by anesthesiologlsts.
Metltods: Data from 5,475 clums ln the Amerlcan Soclety of
alternative drug delivery methods, temporary or long-
Anesthesiologists Closed Clalms Project database between 1970 terin neural blocks, and neuromodulatory techniques'5'6
and 1999 were reviewed to compare ltability related to chronlc The purpose of this study was to identify and describe
paln management with that related to surglcal and obstetric issues and trends in chronic pain management liability
(surglcaVobstetric) anesthesla. Acute pain management clalms
for anesthesiologists. We compared injuries and liability
were excluded from analysis. Outcomes and ltabillty character-
lstics between 2a4 palrl rlnLaroagement clalms and 5,125 surglcaV
characteristics in closed anesthesia malpractice claims
obstetrlc clalms were comPared. arising from chronic pain management with those aris-
Resuhs: Clatms related to chronic pain management ln- ing from surgical and obstetric anesthesia using the
These creased over tlme (P < 0.01) and accounted for 10o/o of all clalms American Society of Anesthesiologists (ASA) Closed
in the 1990s. Compensatory payment amounts were lower in Claims Project database.
numbers chronic pain management clatms than in surgicaVobstetric an-
have esthesla clalms from 197O to 1989 (P < 0.05)' but dudng the
1990s, therewas no dlfference lnslze of payments. Nerve lnjury
risen! and pneumothoftlx wene the most collmon outcornes in lnva- Materials and Methods
slve paln management cl,atms. Eptdural steroid lniections ac-
The ASA Closed Claims Project is a structured evalua-
counted fot 4oo/o of all chronic paln management clalms. Ser{-
ous injudes, involving brain damage or death, occurred wlth tion of adverse anesthetic outcomes collected from the
eptdural sterold lnjectlons wlth local anesthetlcs andlor opl- closed malpractice insurance claim files of more thar:^ 35
otds and with malntenance of tmplantable devices. professional liability companies from throughout the
Concluslons: Frequency and payments of clatns assoclated United States. A detailed description of the data collec-
with clrronlc pain management by anesthesiologists lncreased
in the 1990s. Braln damage and death were associated with tion process for the ASA Closed Claims Project was
epidural sterold iniectlon only when optolds or local anesthet- reported previously.T'8 In brief, a claim file typically
lcs were included. Anestheslologists lnvolved in home care of includes narrative statements from the personnel in-
patients with lmplanted devlces such as morphine pumps and volved, medical records, expert and peer reviews, del>
epidural iniectlons or patlent-controlled analgesla should be osition sufirmaries, outcome reports, and the cost of
aware of potentlal complicatlons that may have severe
settlement or jury award. Anesthesiologist-reviewers
completed a detailed data form plus a narrative summary
for each claim in which the sequence of events and
CHRONIC pain is one of the most common and chal-
nature of injury could be determined from the informa-
lenging medical problems facing our society.l The spe-
tion available in the file. Claims for damage to teeth and
cialty of pain management has grown steadily in recent
dentures were excluded from data collection. All 86
years, largely because of the recognition that multiple
anesthesiologist-reviewers were in active anesthesia
factors contribute to chronic pain.2 Anesthesiologists
practice; 62 were affiliated with academic institutions
have provided leadership in the development of the (38 academic practice and24 pivate practice plus teach-
practice of pain management, with the application of
ing affiliation) and 24 werc in private practice. The
nerve blocks and other technical procedures that are
reviewers followed detailed instructions in completing
firmly linked to a biomedical model of pain.3'a Anesthe-
the data collection form and had access to reviews of
plaintiffand defense pain management experts' The data
$ Ctinical Professor, ll Assisunt Professor,
' Associate Professor,
* Professor and form includes patient characteristics, surgical proce-
Chair, Department of Anesthesiology; f Research Associate Professor, Depart-
ments of Anesthesiology and Anthropology (adiunct), { Professor, Departments
dures, sequence and location of events, critical inci-
of Anesthesiology and Neurological surgery (adiunct)' dents, clinical manifestations of injury, appropriateness
R€c€ived from the Department of Anesthesiology, University of Vashington, of anesthesia care, and outcome. Severity of injury is
Seattle, wNhington. submitted for publication February 28, 2oo3. Accepted for
publication June 16, 2OO3. SuPported in part by the American Society of Anes assigned by the on-site reviewer using the insurance
ihesiologists, Park Ridge, Illinois. All oPinions expressed are those of the authors industry's l0-point scale, ranging from 0 (no apparent
and do not reflect the policy of the American society of Anesthesiologists'
injury) to 9 (death). Severity scores were grouped into
Address reprint requests to Dr. Fitzgibbon: DePartment of Anesthesiology,
Box 35654o, Univesity of wshington School of Mcdicine, 1959 NE Pacilic three broad categories for analysis: temporary and non-
Street, Seattle, washington 981954540. Address electronic mail to:
disabling (score 0-5), permanent disabling injuries
dermot@u.washington.edu. Individual article reprints may be purchased through
the Joumal Web site , ww.mesthesiology.org (score 6-8), and death (score !). Appropriateness of

Anesthesiology, V lO0, No l,Jan2OO4 98


care was rated as standard (appropriate), substandard, or axial (paravertebral, nerve root transsacral); peripheral
impossible to judge on the basis of reasonable and pru- nerve (intercostal, suprascapular, ilioinguinal, genitG
dent practice at the time of the event. The reliability of femoral and others); autonomic (stellate ganglion, celiac
reviewer assessments of appropriateness of care has plexus, lumbar sympathetic); upper extremity @rachial
been found to be acceptable.9 Payment amounts are plexus, other); lower extremity (lumbar plexus, sciatic,
recorded in actual dollars paid, unadjusted for inflation. other); and head and neck (occipital, glossopharyngeal).
Data for this report were derived from the current ASA Pain injections were defined as use of a needle tech-
Closed Claims Proiect database of 5,475 claims for nique intended to reduce inflammation around a nerve
events occurring between l97O and 1999. These consti- or neuraxis or to reduce myofascial, tendon, or joint
tute all claims collected through December 2000. pain. Examples of injections included epidural steroids,
For the current analysis, claims were divided into three trigger point injections, dry needling, botulinum toxin,
groups: acute pain management, chronic pain manage- and tendon or joint injections. Claims involving use of
ment, and surgicaVobstetric anesthesia. Claims associ- steroids plus local anesthetics in the epidural space were
ated with treatment of pain in the immediate postoper- classified as injections (rather than pain blocks), assum-
ative period were included in the acute pain ing that the primary intent was reduction of inflamma-
management group. This group was excluded from anal- tion by steroids, For analysis ofthe role ofsteroids alone
ysis because of the small number of claims. Claims re- uersus steroids with local anesthetics or opioids in the
lated to treatment of nonoperative, nonobstetric pain injection, an injection was classified as "steroid only" if
were included in the chronic pain management group if there was no listing of additional agents (ocal anesthetic
the complaint involved diagnosis or direct complications or opioid) on the data form or in comments in the
of the treatment of chronic pain. The chronic pain man- narrative summary.
agement group also included claims for complications Ablatiue procedures were defined as invasive proce-
arising from the postdischarge treatment of pain result- dures that permanently interrupt neryous system activity
ing from a previous surgical procedure. The third group and were grouped by the agent (alcohol or phenol) or
(surgicaVobstetric) consisted of all other claims in the technique (radiofrequency, cryoanalgesia) used for
database, including anesthesia for surgery or obstetric ablation.
delivery (including labor analgesia). In addition, the third Implantation or remoual of deuices included compli-
group included complications of general anesthesia, cations associated with either the insertion or removal of
monitored anesthesia care, or sedation (such as airway implantable pumps, catheters, and nerve stimulator de-
management problems) during pain management inter- vices. Implantable pumps included devices that contin-
ventions if there v/ere no direct complications of the uously or intermittently deliver medications into the
pain management intervention itself. neuraxis. Catheters included those that were in place for
Chronic pain management claims were divided into days or longer to continuously or intermittently deliver
two main categories based on the mode of treatment: medications into the neuraxis or nerye plexus. Nerve
inuasiue procedures and noninuasiue pain manage- stimulator devices include dorsal column or peripheral
ment. Inuasiue procedures included nerve blocks, injec- nerve stimulators that were surgically implanted.
tions, ablative procedures, implantation or removal of Maintenance of deuices included postinsertion main-
devices, and maintenance of devices (including cathe- tenance, programming, or refill of arry type of neuraxial
ters). Some claims involved multiple interventions or pump, nerve stimulator, or catheter.
other miscellaneous interventions that were not in the Otber inaasiue procedures included the iniection of
categories listed above. If multiple treatments occurred, blood or saline into the epidural space and neurolysis of
the claim was categorized according to the treatment epidural adhesions (either by injection of hlpertonic
that was implicated in the complaint and alleged injury. saline or by the use of specialized epidural catheters).
If it could not be determined which treatment caused Noninuasiue pain management was defined as diag-
the injury, the claim was assigned to a category of "mul- nostic or therapeutic activities that did not require the
tiple procedures." use of needles or the insertion of catheters or devices.
Pain blocks were defined as invasive procedures de- These included primarily systemic medication mxnage-
signed to temporarily interrupt nervous system activity. ment and medical opinions or consultations. Behavioral
Temporary interruption of the nervous system was modification therapy was also considered in this
achieved by injection of local anesthetic agents. Other category.
agents such as steroids may be injected simultaneously, In the ASA Closed Claims Project database, multiple
but the procedure was classified as a block if the primary outcomes were recorded for some claims (for example,
agent was local anesthetic. Blocks intended to tempo- headache plus back pain). These outcomes were as-
rarily intemrpt the nervous system were classi-fied ac- signed by anesthesiologist-reviewers from a preset list
cording to the anatomic site of injection. Anatomic sites provided on the data collection form. For pain manage-
of injection included neuraxial (epidural, intrathecal); ment claims, two of the authors @.R.F,, a board-certified

Anesthesiology, V 100, No 1,Ia:n2OO4

'. I'

pain management physician, and K'L.P.) used content 96of clrim! h lroup

analysis of narrative summaries to develop additional

outcomes not assigned to other claims in the ASA Closed
Claims Project database. These additional outcomes in-
cluded infection, retained catheter or fragment, and a
category for increase or no relief of pain after treatment.
These outcomes v/ere not previously observed in the
overall database in sufficient numbers to warrant inclu-
sion on the data collection form. EA

N.fv. Pnaumo- Hddrdra Bad(Pdn Ertn D.att
Statistical Analitsis Dilr'ragG lhorex Ddntc
For comparison of chronic pain management claims
with surgical/obstetric anesthesia claims, differences in Ffg. 1. Prirnary outcome in chronlc patn aranagetn€nt clalos (sodd
barc) uetsus surgtcavoh€tdc dlalm's (open bats)- ? = 0.05.
proportions were tested for statistical significance with
the Z test.ro Age differences were tested by t test. Pay-
which the event leading to the claim occurred, payments
ment amounts were compared for differences in their
inl97o-1989 chronic pain management claims (median
distribution by use of the Kolmogorov-Smirnov test, with
payment, $25,500) were lower than payments in surgi-
twetailed exact (pennutation) tests to determine statis-
tical significance at a value of P = O.05.rr
callobstetric claims (median payment, $110,000). Pay-
ments in chronic pain management claims were higher
in the 1990s than payments in chronic pain management
claims :ri,l97O-1989 (P = 0.O5). During the 1990s, there
was no difference in size of payments between the
A total of 284 claims were associated with chronic pain chronic pain management and surgical/obstetric claims
management and 5,125 with surgical/obstetric anesthe- (fle. 2).Nearly one third (3Oy") ot chronic pain manage-
sia. Acute pain management, including blocks and intra- ment claims resulting in payment in the 1990s involved
venous patient-controlled analgesia, accounted fot 66 a permanent and disabling iniury, an increase from 17%
claims. The numbers of these claims were insufficient for in 1970 -1989 (not statistically significant). There was no
analysis, so they were excluded from further analysis. difference in the proportion of chronic pain manage-
Claims related to chronic pain management increased ment claims with payment across decades.
over time, accounting fot 2% and 3% of claims in the In nearly two thirds (64%) of chronic pain manage-
1970s and 1980s, respectively, and lO% of all claims in ment claims, the injury became apparent after discharge
the 1990s. Patients in chronic pain management claims from the treatment facility, in conffast to surgical/obstet-
were older than those in surgical/obstetric anesthesia ric claims in which the injury became apparent during
claims (mean, 48 1- 15 [SD] compared with 4l
+ 2o yr, anesthesia care in most claims (83%, P < 0.01; table 1).
P < 0.01). There were no pediatric chronic pain man- The injury was iudged preventable by better preanes-
agement claims and 484 pediatric surgicaVobstetric thetic,/preprocedure evaluation rn 7% and preventable
claims rVomen accounted for 60% of chronic pain
(9%o). by better postoperative/postprocedure care in 12% of
management claims and59% of surgicaVobstetric claims' chronic pain management claims, percentages similar to
Most chronic pain management claims resulted in tem- those in surgical/obstetric claims (table 1)' Documenta-
porary or nondisabling injuries (76% compatedwith49% tion of appropriate informed consent was similar in
of surgical/obstetric claims). Nerve injury' pneumothG chronic pain and surgicaVobstetric claims (table 1)'
rax, headache, and back pain were more common in
chronic pain management claims, whereas death and
brain damage were more frequent in the surgical/obstet- Procedures Leading to Cbronic Pain Management
ric claims (P < 0.05, fig. 1)' The proportion of perma- CInims
nent and disabling injuries (score, 6 - 8) was similar in Of the 284 chronic pain management claims, 276
(97Yo) were for invasive procedures. Blocks and injec-
chronic pan (2O%) and surgical/obstetric claims (l9yo).
Anesthesia care was more likely to meet standards and tions together accounted fot 78% of claims related to
less likely to be substandard in chronic pain claims invasive pain management (table 2). Epidural steroid
compared with surgical/obstetric claims (P < 0.01, table injections (+ local anesthetic and/ot opioids) accounted
1). Compensatory payments were made in more than for 83% of injections and 40% of all chronic pain man-
half of all claims (chronic pain management and surgical,/ agement claims, Peripheral blocks and autonomic blocks
obstetric, table 1). Payment amounts were greater in the each accounted for 36% Q2% total) of the 78 block
surgical/obstetric claims than the chronic pain claims claims (table 2). Of the 20 claims associated with main-
overall (P < 0.01, fi9.2). When analyzed by decade in tenance of devices, 10 were associated with epidural

Anesthesiology, v 100, No l,Jan2Oo4


Table 1. payment, Standard of Care, and Preventlon: Chronic Paln Managementversus Other Clalms
Chronic Pain Surgical/Obstetric
(n = 284) (n : 5,125)


142 53 2,777 59 NS
Payment made to Plaintiff
126 47 1,891 41 NS
No payment
155 65 2,501 56 < 0.01
Standard care <
84 35 1,934 44 0.01
Substandard care
71 36 2,166 83 < 0.01
lniury became apparent in anesthesia facility <
127 64 443 17 0.01
lnjury became apparent after discharge
Complication preventable by better preanesthetic evaluation 15 7 395 I NS
213 93 4,080 91 NS
Not preventable by better preanesthetic evaluation
26 12 431 1'l NS
Complication preventable by better postanesthetic care
88 3,592 89 NS
Not preventable by better postanesthetic care 195
141 66 2,404 72 NS
Appropriate informed consent documented
74 34 959 29 NS
Appropriate informed consent not documented

Claims in which items could not be assessed were excluded from analysis on an item-by-item basis. P values
were calculated by z test.
NS = not statistically significant.

catheters, 7 with implanted pumps, 2 with patient-con- Noninvasive procedures leading to claims included
trolled analgesia devices, and 1 with a plexus infusion. medication prescription or management (five claims),
The most cornmon outcomes for all invasive proce- diagnosis (two claims), and cupping (one claim' table 2).
dures were nerve injury and pneumothorax (table 3). Two of the five claims for medication prescription or
Outcomes dffiered by type of procedure. The most com- management alleged addiction resulting from treatment'
mon complication of blocks was pneumothorax, ac- Another involved overdose and death in a patient who
counting for 5l% of all block claims ltable 3)' Pneumo- had not revealed a previous addiction to opioids' There
thorax was also the most common outcome of trigger Table 2. Procedures tn Chrontc Paln Management Clatms
point and other nonepidural injections' The most com-
mon outcomes of injections inYolving epidural steroids
(+ local anesthetics and/ot opioids) were nerve injury,
infection, and headache (table 3). Nearly hralf (47%) of
the claims associated with ablative procedures involved
lnvasive procedures 276 97
unintentional nerve injury (table 3). Implantation or re- lnjections 138 49
moval of devices more often resulted in infection or Epidural steroids I associated agents 114
retained catheter fragments. The most cornmon out- Trigger point 17
come of claims related to maintenance of devices was Facet
death or brain damage (45%, table 3)' Blocks 78 27
Peripheral 28
t450p6 Stellate ganglion 19
lro.od! Other autonomic I
Neuraxial 9
$350,00 Upper/lower extremity 7
s3op6 Axial 4
Head and neck 2
3260,ru Ablative procedures 17
t20,0& Agent 13
Technique 4
lmplantation or removal of devices 12
3rs,00 lmplantable pumP 5
Nerve stimulator 4
t50,ofl, 3
t0 Device maintenance 20 7
Other interventions' 11 4
Noninvasive pain management 8 3
Ftg. 2. Median payment over dlfferent tlme perlods. Bar hetghts Medication prescription 5
lndtcate medtan paynen:t (sotldDars: chronlc p ln clalmsi open Opinion/diagnosis 2
bars: trglcal/obstetdc cl,atnrs); llnes Trrdlcate 25th and 75th Cupping procedure 1
percentile payment ranges. Payment tn chrontc paln manage-
hent clalms was lower than payment ln surglcaVobstetric Total does not sum to 100% because of rounding.
clalms in 1970-1989. Payments between these groups dld not
dlffer ln lgg}-1r999. 'P < 0.01 between paln management and 'lncludes three claims invoMng muhiple procedure associated with complications.
surgicaVobstetric Payments. One of these claims invofued invasive plus noninvasive pain managarent.

Anesthesiology, V 100, No l,lan2OO4

Note: By definition, Nerve injury claims include Arachnoiditis and SND (severe excruciating
pain from sensory nerve disruptions). Infection claims include spinal Meningitis. Increased
pain/no relief claims overlap SND, or severe Sensory Nerve Disruptions. (see DM package insert.)

Table 3. Prtmary Outcome for Invaslve Paln Management Clatms

lnjections (n = 138)

Epidural Trigger,
All lnvasive Steroid + Facet lmplanV
Procedures Blocks Agents Other Ablative Removal Maintenance Other/MultiPle
(n = 2761 (n = 78) (n = 114) $=2a\ (n = 17) (n = 12) (n:20) (n = 11)

Outcome No.

Nerve iniury 63 23 14 18 28 25 28 847 217 420 545

59 21 40 51 00 18 75 16 00 00 00
lnfection 35 13 2 3 24 21 00 00 325 420 218
Death/brain damage 269 4 5 98 00 16 00 945 327
Headache 21 I 1 1 20 18 00 00 00 00 00
lncreased pain/no
relief 21 I 7 I 10 9 00 00 217 15 19
Retained catheter 93 44 14 00 325 00 00
1 1
'l 44 00 16 18 00 00
Other 42 15

I 10 18 16 313 635 18 420 218

point, facet, and others, are listed separately, with percentage
Epidural iniection of steroids (+ local anesthetics and opioids) and injections, including trigger
percentage in each invasive procedure group. Totals sum to more than 1 00olo
shown for each separate category. otherwise, the percentage of claims implies the
because of multiple complications in some claims.

were two claims for failure to diagnose by patients who ln 34 (58YA, the pneumottrorax was diagnosed after the
consulted with the defendant anesthesiologisr but fol- patient had left the pain treatment facility. In 15 of those
lowed up with another provider. cases, the patient presented at an emergency department
for diagnosis and treatment.
Most Common Complications of Inuasiue Pain Infection was cited in 13% of all claims arising from
Management Procedures invasive pain management procedures. Most of these
Nerve injury was the most common complication of infections were associated with epidural steroid iniec-
invasive pain management procedures. HaIf (n: J2) ot tions (table 3). lthe most common infections associated
these 63 nerve iniury claims involved spinal cord injury. with these injections were meningitis (n : l'2), epidural
These included 14 after epidural steroid injections (6 abscess (n : D, and osteomyelitis (n : 3). Two claims
resulting in paraplegia, I quadriplegia), 5 after blocks (2 involved both meningitis and epidural abscess, with one
with paraplegia), 3 after ablative procedures (1 with of the fwo also involving lumbar osteomyelitis. Six of the
paraplegia), 1 after cervical facet injection, 2 after im- seven epidural abscesses required surgical drainage; one
plantation or removal of devices (1 paraplegia, 1 quad- of the seven resulted in permanent lower-extremity mo
riplegia), 4 after device maintenance (4 with paraplegia), tor deficits. Other infections were associated with im-
and 3 after other invasive procedures (2 with paraple- plantation, removal, or maintenance of implanted de-
gia). Of the 18 claims for paraplegia or quadriplegia, 4 vices (table 3).
were associated with epidural abscess, 8 with chemical Death or brain damage resulted from epidural steroid
injury in which the anesthetic or neurolytic agent was injections (n : 9) and device maintenance 1n : 9).
injected into the spinal cord, and 4withhematoma' Two Iniection of opioids, local anesthetic, or both occurred
of the claims for hematoma involved administration of in 6L% of the 114 epidural steroid injection claims.
epidural steroids in patients who received anticoagu- Death or brain damage occurred only in epidural steroid
lates. Other nerve injuries associated with invasive pain
injection claims that involved local anesthetics with or
management procedures included lumbosacral nerve
without opioids in the injection (fig. 3). This dffierence
root (n : 2l), sciatic nerve (n : 2), and brachial plexus
(n:2). in severe outcome (death or brain damage) between
epidural steroid injections with uersus without associ-
Of the 59 claims for pneumothorax, 40 involved pain
ated agents was statistically significant. Six of the severe
blocks and 18 involved iniections' Of the 4O pan blocks
injuries occuffed shortly after local anesthetic adminis
leading to pneumothorax, 23 were intercostal, 8 stellate
ganglion, I suprascapular, 2 supraclavicular brachial tration with either an unintended intrathecal injection
plexus, 2 interscalene brachial plexus, and 2 were thoracic
(n : 5) or allergic reaction (n : 1). A documented test
paravertebral. Of the 18 claims for pneumothorax associ- dose was given in only two of these six patients. In one
ated with injections, 15 were trigger point, and 1 each was patient, cardiovascular collapse and respiratory depres-
costochondral, thoracic facet, and interscalene botulinum sion developed after inadvertent intrathecal injection of
toxin. In 3l (fiyA of the 59 pneumothorax claims, it was 6 ml of local anesthetic during attempted thoracic epi-
explicitly stated that chest tubes were placed for treatment' dural steroid injection. Three severe outcomes were the

Anesthesiology, V 100, No 1,Jan2OO4


i6 of claima ln grouP and/or opioids, maintenance of implantable devices, and

procedures complicated by a pneumothorax.

Metbodologic Issues
Before the results can be interpreted, it should be
emphasized that closed claims analysis has a number of
well-described limitations.s The Closed Claims Proiect
database can provide only an indirect assessment of the
safety and liability risks of anesthesiology-based pain
management practice in the United States. In particular,
r$xvcrnJury lnfccdon nradrchc ogflffi" it cannot estimate the relative frequency of claims in
iiy*l$, pain management compared with other areas of anesthe-
in epidural tniectlons. Soldd sia because of the lack of denominator data. Closed
Fig. 3. Most cofirmon outcomes
&als represent tnjecttons wtth steroids only. Op en bars lndlcate claims are biased by the presence of more severe and
lniectlons ln which local anesthetlc or oplold (or both) were costly injuries, because plaintiff attorneys are unlikely to
added to the steroid. *P < 0.05 between proportlon of lniectlon pursue claims with an estimated financial recovery for
group with that outcome.
damage of less than $50,000.12 Other limitations of the
analysis of closed claims include the absence of rigorous
result of a delayed respiratory depression from epidural
comparison groups and partial reliance on data from
morphine administered along with the steroid.
direct participants rather than impartial
Death (n - 4) or brain damage (n : 5) associated with
Claims spanned a period of time during which practice
maintenance of devices involved implanted pumps (n :
patterns changed. The analysis also evaluated only the
4), epidural injections (n : 3), and patient{ontrolled
information in the database that was transcribed to the
analgesia (n : 2). All nine claims involved the use of
data sheet by the reviewer, who depended on the infor-
opioids (morphine [n : 8], hydromorphone [n : 1])'
mation contained in the insurance company file. Spe-
and most (eight of nine) involved administration of the
cffic, detailed information regarding signs and mecha-
wrong dose of opioid. In four patients, the overdose
nism of injury may therefore be incomplete compared
arose from a pump programming effor when the con-
with a prospective study. The retrospective case review
centration of opioid was changed. The programming
studies included in the database were also selected in a
errors involved either a failure to recalibrate with a more
nonrandom fashion, without control over geographic
concentrated solution of morphine (n : 2) or recalibra-
balance. Closed claims analysis for pain management is
tion for lower concentration of morphine, with failure to
useful for generating hypotheses about the mechanism
recognize greater strength of solution that remained in
and prevention of pain management injury, but it cannot
the catheter (n : 2). The other cases of overdose in-
be used for testing those hypotheses. As a retrospective
volved drug interactions with concomitant use of other
study, it cannot establish a cause-and-effect relationship
central nervous system depressants (n : 2) or intrathe-
between previous events or between changes in claim
cal migration of an epidural catheter (n : 2). One obese
patient had sudden cardiac death while on appropriately
dosed morphine patient-controlled analgesia, with au-
Cbronic Pain Management Liability
topsy revealing severe cofonary artery disease. seventy-
Given the prevalence of pain management, particularly
eight percent (seven of nine) of the claims for death or
during the 1990s, it is not surprising that the overall
brain damage related to maintenance of devices were
percentage of chronic pain management claims has in-
judged by reviewers to represent substandard carc' Pay-
creased from2-3% in the 1970s and 1980s to 10% in the What is the
ments were made in eight of nine of these claims, and
1990s, when they were on a par with obstetric anesthe- percentage
the payments were high (median, $3O9,129; nnge, today
sia claims. The majority of pain claims involved invasive
$29,999 to $ 1,600,000).
procedures such as blocks, injections, ablative proce- considering
dures, and insertion and/or removal of implantable the massive
pumps of stimulators, probably reflecting the risks in- growth in
herent in these techniques compared with those inher- the last 10
Claims arising from chronic pain management in- ent in medical management. Nerve damage, pneumothG
creased over time, accounting for lO% of anesthesia rax, headache, and back pain were more common in
malpractice claims in the 1990s' Although nerve injury chronic pain management claims, whereas death and
and pneumothorax were the most coflrmon outcomes in brain damage were more frequent in the surgicaVobstet-
pain management claims, serious injuries involving brain ric claims (fig. 1). However, the proportion of perma-
damage and death occurred. Specific areas of concern nent and disabling injuries were similar in chronic pain
include epidural steroid iniections with local anesthetics and surgical/obstetric claims, demonstrating that signif-

Anesthesiology, V 100, No l,J^n2OO4


icant injury can also occur with pain management inter- Interestingly, death or brain damage was observed in
ventions. Nerve injury occurred in 23% of pain manage- our study only when local anesthetics or opioids (mor-
ment claims, and death or brain damage occurred in lO% phine) were injected concomitantly with the epidural
(fig. 1)' steroid. Failure to appropriately manage cardiovascular
Although overall payment in chronic pain claims depression from epidural or intrathecal local anesthetics
seems to be lower than surgical/obstetric claims, there or delayed respiratory depression from epidural mor-
was no difference in payments made to the plaintiff for phine accounted for these poor outcomes. Therefore,
chronic pain claims and surgical/obstetric claims in the ASA Closed Claims Project data demonstrate that serious
1990s (fig. 2). The trend toward increasing payment injuries can occur with epidural steroid injections when
amounts in pain management claims is particularly im- combined with local anesthetics and opioids. However,
portant because payment totals in pain claims generally we are unable to determine the incidence of injuries
are made on behalf of a single defendant and therefore with epidural steroids because of the limitations of the
do not include payments made on behalf of other prac- closed claims methodology. Because of the lack of de-
titioners (e.g., surgeons and obstetricians) or hospitals nominator data, it is not clear whether these injuries are
that are often part of the total payment amount in the a result of more blocks being performed, a more litigious
surgical/obstetric claims in the Closed Claims Project population, or other factors. Our data suggest that pa-
database.T Because payment is influenced primarily by tient safety may be improved by excluding typical epi-
severity of injury,7 these data may reflect an increasing dural doses (volumes in excess of intrathecal test doses)
severity of injury from chronic pain management claims. of local anesthetics and/or opioids from epidural steroid
This is of concern to pain management anesthesiologists, injections.
particulady as newer and more invasive procedures are
developed. Irnplantable Deuices
Complications related to implantable devices occur and
Epidural Steroid Ini ections may place patients at risk for injury.17 In a prospective
Epidural steroid injections have been used to treat study, Follett and Naumannrs noted that ttre frequency of
spinal and radicular pain for more than 40 yr, with the procedure-related complications underscored the need for
majority of injections performed by anesthesiologists.
physicians performing implantations to use carefirl surgical
Epidural steroid injection is reputed to be a very safe technique and follow implant guidelines. In our study of
procedure with a very low infectious complication surgical-type procedures (implantation/removal of de-
rate.ta Abram and O'Connor15 reviewed complications vices), complications such as nerve iniury (17%o) or infec-
associated with epidural steroid iniections under the tion (25'{/o> featured prominently (table 2). Clearly, when
headings neurologic dysfunction (arachnoiditis, aseptic anesthesiologists perform procedures in which issues such
meningitis, others), infections (meningitis, epidural ab- as infection, bleeding, nerve injury, or pneumothorix may
scess), steroid side effects and complications, technical potentially occru, management strategies should be incor-
complications (postdural puncture headache, hema- porated into after-procedure care plans. Although inflam-
toma), and minor side effects. The authors concluded matory mass lesions have been associated with long-term,
that there was little risk of serious complications associ- highdose, and highconcentration intrathecal infusions of
ated with the use of epidural steroid injections. In par- morphine and hydromorphone,re this complication was
ticular, aseptic meningitis and bacterial meningitis not observed in our series.
seemed to be uncommon but real risks. In contrast, our Advances in technology, particularly during the 1990s'
study found that 4O% of all chronic pain claims were such as implantable therapy (infusion pumps, spinal cord
associated with the injection of epidural steroids, which stimulators) and ambulatory infusion devices, have facili-
may reflect the frequency of these procedures. Serious tated the tfansition of traditional hospital ffeatments of pain
infectious complications (epidural abscess, meningitis' to the home environment.2o-23 Some authorszt'24 ytora
and osteomyelitis) were observed in 20 of 114 epidural suggested that opioid administration (neuraxial, patient-
steroid claims. Hypothetically, injection of steroids into controlled analgesia) may be achieved effectively and safely
the epidural space may result in local immunosuppres- in the home environment. In our analysis of 20 claims
sion with possible infectious consequences in predis- associated with maintenance of devices, 45% were associ
posed individuals. In addition, significant nerve injury ated with death or brain damage when events such as
(with seven claims of quadriplegia/paraplegia) was also pump programming e(rors, drug overdose, and concomi-
observed in 28 epidural steroid claims. Two of the three tant use of other central nelous system depressants were
claims for hematoma from epidural steroids were in implicated in the injury. This might suggest an area for
patients who received anticoagulates. Likewise, claims increased awareness of potential problems for anesthesiol
for spinal cord injuries increased in the 1990s, perhaps ogists involved in home care of patients with implanted
because of neuraxial blocks in patients who received devices such as morphine pumps and epidural iniections
anticoagulans.l6 and with patient{ontrolled analgesia.

Anesthesiology, V 100, No l,Jan2OO4


cisco, California; Pennsylvania Medical Society Liability Insurance Company'

Pneumothorax Harisburg, Pennsylvania; Physicians Insumnce A Mutual Company, Seattle,
Twenty-one percent of claims were related to pneu- Washington; Physicims Insumnce ComPany of Wisconsin' Madison, Wisconsin;
Preferred Physicians Mutual Risk Retention Group, Mission, Kans6; State Volun-
mothorax, which was associated primarily with intercos- teer Mutual Insumnce Company, Brentwood, Tennessee; University of Texas
tal nerve blocks, trigger point iniections, and stellate Medical System, Austin, Tex6; Utah Medical Insurance Association, Salt lake
City, Utah; and Vetemns Adminisration, washington' D.c.
ganglion blocks. Of the 59 claims for pneumothorax,
many involved delay in diagnosis and treatment andlor
insertion of chest tubes for treatment. The actual occur- References
rence of these aggravating factors in pneumothorax 1. Gallagher RM: chronic Pain: A public health problem? clinJ Pain 1998;
claims may be higher, because these data were not 14:277-9
2. Rowbotham DJ: Advances in pain. BrJ Anaesth 2001; 87:1-2
itemized on the data collection form. There may be 3. Bacon DR, Reddy V, Murphy OT: Regional anesthesia and chronic pain
additional claims with delayed treatment or chest tubes management in the 1920s and 1930s: The influence of the American Society of
Regional Anesthesia. Reg Anesth 1995i Z0tlal-gz
in which the on-site reviewer did not explicitly state 4. Bridenbaugh PO: 1994 Gaston Labat Award lecture. Anesthesiology and
these factors in the narrative. The occurrence of iatro pain managemmt: Medical practice or perception Reg Anesth 7994; 19:501-6
5. Ashbum MA, Staats PS: Management of chronic pain. Irncet 1999" 353:
genic pneumothorax may be associated with significant 1865-9
morbidity and mortality,2s with some authors recom- 6. Cousins MJ: Pain: The past, present, and future of anesthesiology? The E' A'
Rovenstine Memorial lf,cture. ANETrlsIormv 1999; 9l:.53a -51
mending chest tube drainage if the volume of the pneu- 7. Cheney FW, Posner K, Caplan RA, Ward RJ: Standard of care md anesthesia
mothorax is greater ttranlow of the pleural space.'6 liability. JAMA ; 261 :1599 - 6O3
8. iheney Fw: '9a9The American society of Anesthesiologists closed Claims
Although pneumothorax is a recognized complication of Proiect: What have we learned, how h6 it affected pmctice, and how will it
proced.rres such as intercostal,2T brachial plexus,28 and affect practice in the future? AmrlslolmY 1999,91:552-6
9. Posner KL, Sampson PD, Caplan RA, Ward RJ, Cheney Fw: Measuring
stellate ganglion blocks,2e the occurrence of pneumo- inteffiter reliability among multiple mters: An example of methods for nominal
thorax for procedures such as trigger point iniections data. stat Med 1990; 9:1103-15
10. FleissJl: Statistical Methods for Rates and Proportions, 2nd edition' New
may be somewhat surprising to many anesthesiologists, York, Wiley and Sons, l9al, PP 29-30
because this complication is not ofren reported with this I 1. Mehta CR, Patel NR: SPSS Exact Tests 7.0 for windows. chicago, SPSS Inc'
procedure.3o Our data suggest that it is important to 1996, pp 75-9r
12. Huycke LI, Huycke MM: Chamcteristics of potential plaintiffs in malpmc-
establish a monitoring system for pneumothorax and to tice litigation. Ann Intern Med 1994; 720t792-a
instruct patients as to the symptoms and signs of a 13. CadnoJA, Morrison WB, Parker L, Schweitzer ME, Levin DC, SunshineJH:
Spinal iniection procedures: Volume, provider distribution, and reimbursement
pneumothorax after intercostal nerve blocks, stellate in the U. S. Medicare population from 1993 to 7999. Radiology 2002; 225:723-9
14. Lowell TD, Errico TJ, Eskenazi MS: Use of epiduml steroids after discec-
ganglion blocks, trigger point iniections, and brachial tomy may prectispose to infection. spine 2000; 25:516-9
plexus blocks. 15. Abmm SE, O'Connor TC: Complications associated with epiduml steroid
iniections. Reg Anesth 1996; 2l:149 - 62
16. Cheney Fr$(/, Domino KB, Caplan RA, Posner KL: Neile iniury associated
with anesthesia: A closed claims analysis. AwsIoLmY 1999;90:202-7
Conclusion 17. Levin AB, Sperling KB: Complications associated with infusion pumps
implanted for spasticity. Stereotact Funct Neurosurg 79951 65:147-51
i8. foll.n KA, Naumann cP: A prospective study of catheter-related compti-
In summary, the proportion of claims and compensa- cations of intmthecal drug delivery systems. J Pain Symptom Manage 2000;
tory payment related to chronic pain management l9:2O9 -75
19. Coffey RJ, Burchiel K: Inflammatory mass lesions rosociated with intmthe-
claims in the ASA Closed Claims Proiect database in- cal drug infusion catheters: Report and obseryations on 41 patients. Neurosur-
creased in the 1990s. Nerve injury and pneumothorax gery 2OO2:50:78-86
20. Aldrete JA: Extended epiduml catheter infusions with analgesics for pa-
were the most common outcomes of invasive pain man- tients with noncancer pain at their homes. Reg Anesth 1997:22:35-42
agement claims. Severe injuries involving brain damage 21. Anderson VC, Burchiel KJ: A prospectiYe study of long-term intmthecal
morphine in the management of chronic nomalignant pain. Neurosurgery 1999;
or death occurred with epidural steroid injections with 44:289-30A
local anesthetics andlor opioids and maintenance of im- 22. Bflzenor GA: Long tem intrathecal administmtion of morphine: A com-
parison of bolus iniection via resetroir with continuous infusion by implanted
plantable devices. pump. Neurosurgery l9a7; 21:4a4 -91
23. Dahm P, Nitescu P, Appelgren L, Curelaru I: Efficacy and technical com-
The authors thank Lynn Akedund for her expert secretarial asistance and plications of tong-term continuous intmspinal infusions of opioid md/or bupiv-
Pauline Cooper, 8.S., and John Campos, M.A., for technical ssistance' They are acaine in refmctory nonmalignant Pain: A comparison between the epidural and
members of the Closed Claims Proiect research staff in the DePartment of the intmthecal approach with extemalized or implanted cathetes and infusion
Anesthesiology at the University of Wchington, Seattle,
rvshington The au- pumps.
- Clin J Pain l99ai l4:4 -16
thos also thutk th. members of the American society of Anesthesiologists who 24. Swanson G, Smith J, Bulich R, New P, Shiffmm R: Patient-controlled
se ryed as reviwers for the Closed Claims Proiect. A list of reviewers
is available analgesia for chronic cancer pain in the ambulatory settinS: A rePort of 117
from the authors. The following 24 organizations gave permission for acknowl- patients. J Clin Oncol 1989; 7:1903-8
edgment as a source of closed claims: Anesthesia Setrices Medical Group, Inc ' 25. Despars JA, smsoon Cs, Ught Rw: signifcance of iatrogenic pneumotho
San Oiego, Califomia; Anesthesiologists Professional Assumnce Company, Miami mces. chest 1994i 105:7147-50
Beach, Flori<ta; Armed Forces Institute of Pathology, Silver Spring' Maryland; Risk 26. Weissberg D, Refaely Y: Pneumothoro: Experience with 1,199 patients.
Management Foundation, Cambridge, Massachusetts; COPIC Insumnce Com- Chest 2000; 777 i7279 - 85
p"ny, D.nn.., Colorado; The Doctors' Company, Napa, California; Daughtes of 27. Holzet A, Kapral S, Heltwagner K, Eisenmenger-Pelucha A, Preis C: Severe
bharity National Health Systems, St. Iruis, Missouri; Illinois State Medical Inter- pneumothoro after intercostal newe blockade: A c6e report. Acta Anaesthesiol
Insumce Exchange, Chicago, Illinois; Medical Professional Insumnce Associa- Scand 1998; 42:1124-6
tion, Boston, Massachusetts; Medical Association of Georgia Mutual Insumnce 28. NealJM, HeblJR, GerancherJC, Hogm QH: Bmchial plexus anesthesia:
Company, Atlanta, Georgia; Medical Uability Mutual Insumnce Company, New Essentials of our cutrent undestandin8. Reg Anesth Pain Med 2O02;27:402-2a
Yorlg New York; Medical Mutual Insurance Company of Maine, Portland, Maine; 29. Thompson KJ, Melding P, Hatangdi VS: Pneumochylothoro: A rue com-
Midwest Medical Insumnce Company, Minneapolis, Minnesota; Mutual Insumnce plication of stellate ganglion block. ANEs'nIsIoLGY 7981: 55:549-97
Company of Arizona, Phoenix, Mzona; National Capital Reciprocal lnsumnce 30. Shafer N: Pneumothorax following "tri88er point" iniection (letter). JAMA
company, rvffihington, D.C.; NORCAL Mutual Insumnce Company, San Fran- 7970:213:1793

Anesthesiology, V 100, No l,J^n2OO4