1. Trang chủ
  2. » Y Tế - Sức Khỏe

THE PEDIATRIC DIAGNOSTIC EXAMINATION - PART 10 pps

83 197 0

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Tiêu đề The Pediatric Diagnostic Examination - Part 10 pps
Tác giả Greydanus DE, Patel DR, Pratt HD
Trường học McGraw-Hill Education
Chuyên ngành Adolescent Medicine
Thể loại essay
Năm xuất bản 2006
Thành phố New York
Định dạng
Số trang 83
Dung lượng 4,16 MB

Các công cụ chuyển đổi và chỉnh sửa cho tài liệu này

Nội dung

This can be quantified by the test’s sensitivity the probability that the test is positive in a patient known to have the condition being tested for and specificity the probability that

Trang 2

The Pelvic Examination 729

and theca lutein cysts Other benign ovarian tumors include dermoids (mature cystic teratomas) and sex cord tumors (thecomas, fibromas, and gonadoblastomas) Fibromas are associated with the Meigs syndrome

(ascites, pleural effusion, and fibrous ovarian tumor), usually seen in

adult females Gonadoblastomas occur in those with gonadal dysgenesis and

a Y chromosome They are at increased risk for a malignant germ celltumor, and the gonadoblastomas may be associated with primary amen-orrhea, virilization, and developmental abnormalities The most com-

mon malignant ovarian neoplasm in adolescent females is the germ cell carcinoma [dysgerminoma (most common), embryonal carcinoma, endoder- mal sinus tumor, polyembryoma, choriocarcinoma, and immature teratoma] Other malignant ovarian neoplasms include the Sertoli-Leydig cell tumor associated with virilization and the granulosa cell tumor associated with iso-

sexual precocity (See Essential Adolescent Medicine, Greydanus DE,Patel DR, Pratt HD, eds New York: McGraw-Hill, 2006, p 608–612.)

Vaginal Discharge TABLE 21–5 reviews various causes ofvaginal discharge in adolescent females Vaginal discharge in an ado-lescent may be an indication of normal estrogen stimulation or geni-

tal infection owing to sexual activity Physiologic leukorrhea develops in

young adolescents because of estrogen stimulation, typically ing some months before menarche and continuing for a number of yearsafter menarche The discharge can be clear to white, watery to mucoid,and scant to copious in nature; there is no odor or irritation Non-sexually active adolescents also may develop a vaginal discharge owing

appear-to vaginal foreign-body retention, group A strepappear-tococcal infection, genital contact dermatitis, candidiasis (Candida albicans, C glabrata), or bacterial vaginosis Sexual behavior also can be involved with the development of

candidiasis, group A streptococcal infection, and bacterial vaginosis.Sexually active adolescent females may develop a vaginal dischargebecause of acquisition of an STD or infection (STI) STD agents include

C trachomatis, N gonorrhoeae, herpes simplex virus, and T vaginalis

Ado-lescents are at increased risk for STDs because of high sexual activityrates, multiple sex partners, use of sex and drugs concomitantly, imma-ture cervix, magical thinking of adolescence (i.e., no harm will come tothem despite high-risk behavior), and difficulty dealing with the med-ical system for treatment TABLE 21–6 considers STDs not reviewed inTABLE 21–5

The Pelvic Examination

Most adolescents have unspoken fears and anxiety regarding the pelvicexamination Clinicians also vary in their degree of comfort, skill, andknowledge of performing this examination Many clinicians use an adultmodel to perform the pelvic examination While the technique may not

be significantly different, the levels of sensitivity, communication, andpatience required for an adolescent are significantly different The pelvicKEY PROBLEM

Trang 3

contraception; adolescent may complain ofyellow staining of underpants.

Retained foreign bodies in the vagina can lead

to a brown or bloody foul-smelling vaginaldischarge

Vaginal discharge that is sometimes bloody; itmay or may not be associated with astreptococcal pharyngitis and sexually activityChemical irritation of the genitalia can lead to asensitization reaction; can see vaginitis,urethritis, proctitis, or dermatitis of the externalgenitalia; precipitants can include bubble baths

or soaps, vaginal deodorant sprays, douches,perfumes, latex condoms or diaphragms,vaginal spermicides, sexual lubricants, andothers

STDs if sexually active;

bacterial vaginosis ifthere are symptoms(as vaginal odor)Group A streptococcalvaginitis

Foreign body vaginitis

Scabies, pediculosis,eczema, tinea cruris,psoriasis, eczema,other types ofdermatitis (seeChapter 16)

Saline preparation ofvaginal fluid (normalresults

Foreign-bodyvisualization

Vaginal culture

Examine for fungi,

scabies, (Sarcoptes scabiei) and pediculosis (Phthirus pubis).

Trang 4

Clinical syndrome owing to the normal flora of

Lactobacillus spp being replaced by anaerobic bacteria such as Gardnerella vaginalis,

Mycoplasma hominis, Mobiluncus spp., Prevotella;

sexually associated disorder; also seen innon-sexually active female adolescent; thin,white, malodorous, nonirritating, andnonpruritic vaginal discharge that clings tovaginal walls

Tinea cruris, chemicalvaginitis, other STDs

if sexually active

Chemical vaginitis,candidiasis; ifsexually active:

trichomoniasis

Wet mount (saline or10% KOH todemonstrate yeast orpseudohyphae); Gramstain; fungal culture(Nickerson medium)Wet mount with cluecells [epithelial cellscovered (“studded”)with many gram-negative bacilli];

“whiff test”: fishyodor before and after10% KOH is added;gram stain; vaginalfluid pH <4.5

(Continued)

Trang 5

TABLE 21–5 Vaginal Discharge in Adolescent Females (Continued)

vaginocervical ecchymosis (“strawberrymarks”); swollen vaginal papillae

STD that can be asymptomatic; cervicitis withvaginal red mucosa, hypertrophic cervicalerosion, and purulent (mucopurulent) cervicaldischarge), PID, perihepatitis (Fitz-Hugh-Curtissyndrome), proctitis, pharyngitis, and others;

infections in males also include urethritis,nongonococcal urethritis (NGU), prostatitis, andepididymitis

Bacterial vaginitis,chemical vaginitis,candidiasis, urethritis,cystitis

Gonorrhea,trichomoniasis,bacterial vaginosis

Mycoplasma genitalium,

U urealyticum

Saline prep (movingtrichomonads), Papsmear, rare:culture

Cell culture, nucleic acidamplification tests orNAATs (i.e.,polymerase chainreaction, ligase chainreaction, others),enzyme-linkedimmunoassay (EIA,ELISA), directfluorescent antibody(DFA), and DNAprobe (Gen-ProbePACE 2 assay)

Trang 6

Cervicitis with a purulent or mucopurulent(yellowish) discharge in the cervix and vagina

sometimes caused by N gonorrhoeae or C.

trachomatis; sometimes no microbe identified.

Chlamydia,

trichomoniasis,bacterial vaginosis,

Mycoplasma genitalium,

U urealyticum

Gram stain of thedischarge (pairs ofgram-negative,kidney-bean-shapeddiplococci); culturewith Thayer-Martinmedium in a 10%carbon dioxideenvironment; DNAhybridizationtechniques (Gen-Probe), and NAATs.Gram stain of thecervical discharge;laboratory testing for

N gonorrhoeae or C trachomatis.

(Continued)

Trang 7

TABLE 21–5 Vaginal Discharge in Adolescent Females (Continued)

hyperesthesia may first develop, followed bysmall-group vesicles on erythematous bases;

these lesions become small, shallow, painfululcers on the genitals along with inguinallymphadenopathy

Viral serology,immunofluorescenttechniques, and culturewith typing areavailable as diagnostictests; glycoprotein G-based HSV-2 enzyme-linked immunoassayalso available; Giemsastain or Wright stain(Tzanck test) of materialcollected from a vesicle

or ulcer will revealballoon cells withintranuclear bodies ormultinuclear giant cells;Pap smear also mayshow the multinucleargiant cells; electronmicroscopy revealsviral herpetic particles

Abbreviations: KOH = potassium hydroxide; NAAT = nucleic acid amplification test; U = Ureaplasma; Pap = Papanicolaou.

Trang 8

TABLE 21–6 Sexually Transmitted Diseases

Due to H ducreyi and characterized by genital

ulcers, especially in males (penile ulcers); starts

as small erosion that is red and becomes apainful ulcer(s); unilateral, suppurative inguinallymphadenopathy

Due to T pallidum; red ulcer called a chancre

develops that is not painful unless complicated

by secondary infection (see Chapter 16);

inguinal lymphadenopathy develops that isnontender and unilateral or bilateral; chancre isthe primary stage; untreated, other stagesdevelop: secondary, asymptomatic, early latent,late latent, late (tertiary), neurosyphilis, andrelapsing (owing to HIV/AIDS)

Due due to Calymmatobacterium granulomatis;

starts as an erythematous papule (nodule) thatbecomes a painless ulcer with granulationtissue; inguinal granulomas may look likeinguinal lymphadenopathy and are called

pseudobubos; rare in the United States.

Syphilis, granulomainguinale,lymphogranulomavenereum, genitalherpes

Chancroid, granulomainguinale,

lymphogranulomavenereum, genitalherpes

Chancroid,lymphogranulomavenereum, syphilis,genital herpes

Rule out other STDgenital ulcers; PCR testfor chancroid

available; chancroidculture if available.Nontreponemal tests(RPR,VDRL),treponemal tests (FTA-ABS, MHA-TP); dark-field examination for

T pallidum.

Rule out other STDulcers; Donovanbodies on tissue smear;biopsy

(Continued)

Trang 9

TABLE 21–6 Sexually Transmitted Diseases (Continued)

Human papillomavirus (HPV); >100 types; oftenasymptomatic; warts or condylomas (seeChapter 16) can be seen on the genitals; see link

of 15 oncogenic types (such as 16, 18, 31, 45,and others) to cervical neoplasia

Scabies, pediculosis,eczema, tinea cruris,psoriasis, eczema,other types ofdermatitis (seeChapter 16)Molluscumcontagiosum,condyloma lata(syphilis), urethralprolapse, skin tags(perianal), and benignpearly penile papules(males)

Rule out other STDulcers; complementfixation

Biopsy; colposcopy;cytology (Pap smear);molecular diagnosticmodalities (in situhybridization, dot-blot[ViraPap/Vira Type]),Southern blot andPCR

Trang 10

to those under age 25; over 50 million humansare infected globally, with 30 million deathsdue to HIV/AIDS.

Mainly seen in homosexuals with oral to anal ororal to oral sex; organisms noted include

Shigella spp., Giardia lamblia, Entamoeba histolytica, and Campylobacter jejuni; can lead to

enterocolitis or proctitis; some have nosymptoms

Pinworms travel to the vagina and cause intensepruritus, especially at night; can be sexuallyacquired; see Chapter 20

Infestation of lice (Phthirus pubis) on pubic hair as

well as perineal and axillary hair; also seen ineyelashes; intense pruritus can develop, leading

to mild to severe skin excoriations; oftensexually acquired but also spread via fomites(bedding or clothes)

Broad differential; acuteretroviral syndrome

of HIV/AIDSdevelops within firstfew weeks afterinfection and presentswith fever, malaise,skin rash, andlymphadenopathyDifferential diagnosis ofproctitis and

enterocolitis

Tinea cruris, chemicalvaginitis, candidiasisScabies, tinea cruris,chemical vulvitis

HIV antibody testing(HIV-1, HIV-2):enzyme immunoassay(EIA) as screening test;confirm with Westernblot (WB) or IFA(immuno fluorescenceassay); plasma HIVRNA

Stool culture and stoolexamination forbacteria and parasites

Stool examination;cellophane tape testPubic hair examinationusing a hand lens tosee the adult formand/or the eggs(“nits”)

(Continued)

Trang 11

TABLE 21–6 Sexually Transmitted Diseases (Continued)

Due to the mite Sarcoptes scabiei; the female

deposits eggs in the skin (stratum corneum),which leads to intense itching, especially atnight; various lesions develop: papules,vesicles, pustules, burrows; especially seen overthe hands (finger webs; see Chapter 16) but alsowrists, axillae, belt line, buttocks, breasts,areolae (females); penis and scrotum in males;

sexually acquired

Poxvirus infection that leads to a variable number

of asymptomatic papules that are white topearl-colored; larger ones can be umbilicated;

can be sexually acquired; found in variousplaces, including genital areas; dermatitis maydevelop around the lesions; lesions maydevelop in areas of atopic dermatitis

Pediculosis, tinea curis,chemical vulvitis

Human papillomavirusinfections, atopicdermatitis, tineacruris

Microscopic examination

of scrapped lesionswith immersion oil tofind the eggs, adultmites, or feces

Trang 12

Tinea cruris Fungal infection owing to Trichophyton

mentagrophytes or Epidermophyton floccosum that

involves crural folds and inner thighs (“jockitch”), usually without infecting the penis orscrotum in males; appears as pruritic, redlesions that are slightly raised and have scalingborders; can be infected by tinea pedis or withinfected underwear

Contact dermatitis(allergic or irritant),erythrasma, (due to

Corynebacterium minutissimum)

candidiasis, intertrigo

Wet mount with 10%KOH; Wood lampexamination; fungalculture

Abbreviations: STD = sexually transmitted disease; PCR = polymerase chain reaction; RPR = rapid plasma reagin; VDRL = Venereal Disease Research Laboratory;

FTA-ABS = fluorescent treponemal antibody absorption; MHA-TP = microhemagglutination assay; KOH = potassium hydroxide.

Trang 13

740 Chapter 21: The Gynecology System and the Adolescent

examination procedure should be the last component of the completephysical examination

Performing an age-appropriate pelvic examination must begin withplacing the adolescent patient at ease Begin with establishing a rapportwhile the patient is fully dressed Addressing the patient’s anxiety, pro-viding knowledge about the female anatomy, and educating the patient

on the procedure in a manner that allows her to ask questions is tial In order to ease patient anxiety, allow the patient to request a femaleclinician or to have her parent or friend in the room for support Reassure the adolescent female that the speculum and other equip-ment are not large This may reduce her fears and allow relaxation ofher pelvic muscles Try to discuss some common interesting subject (i.e.,school, significant other, or hobbies); this may help to defocus the patientfrom the procedure and decrease her tension Having the patient breathdeeply is another method for relaxation

essen-Reviewing her physical findings is important at this point Many lescents are not familiar with their bodies and feel uncomfortable withallowing others to view their genitals and breasts Demystify the femalegenitalia by allowing the adolescent to take part in the examination Onecan do this by having a mirror that the patient can hold and view her gen-italia while the clinician is performing the examination Obtaining knowl-edge of the internal and external genitals and their function allows theadolescent female to gain a better understanding of the signs and symp-toms of disease as well as the importance of preventive health visits The equipment required includes a light source, specula, water-basedlubricant, Papanicolaou smear testing kit, and other diagnostic test mate-rial as indicated by the history The specula are made of metal or plas-tic and often come in two types, Pedersen and Graves They differ inthe width of the blades; the Pedersen is generally narrow and best forvirgins, younger adolescents, and those with a narrow introitus Theexaminer should have knowledge of the mechanism and function of thespeculum prior to performing the examination, knowing how to openand close the blades as well as how to lock and release them This willcontribute to diminishing the patient’s anxiety

ado-Prepare and position the patient on the examining table once sheundresses and is properly gowned in private Always keep the patientappropriately covered The clinician should remind the patient of the steps

of the procedure and the reason for the examination The table’s stirrupsshould be in proper position Ask the patient to place her heel into thestirrups, and then ask her to slide down to the edge of the table such thather buttocks are slightly over the edge, providing reassurance that shewill not fall The table may be slightly elevated and the examiner mayinsert a pillow at the back of her head for comfort Ask her to rotate herthigh externally with abduction and flexion (lithotomy position) The two main components of the pelvic exam are the internal andexternal parts

External

||

Internal Speculum Bimanual

Trang 14

Examination of the external genitalia begins with the gloved cliniciansitting in a manner that allows him or her to be comfortable and haveadequate visualization of the pelvis and the vulva

Inspection

Do a visualization and assessment of the pubic hair distribution andcharacter over the mons pubis, lower abdomen, and inner thigh Recordthis using the Tanner stages (see Chapter 13) Inspect the labia majora,labia minora, clitoris, urethral meatus, and introitus During inspection,make note of anatomic abnormalities or variation, including evidence

of masculinization, inflammation, excoriations, papules, discharge,imperforate hymen, or cystic nodules

pres-Speculum

Angle the speculum at about 15 degrees, sliding it over the middle ger; remove the finger Once past the sensitive urethra, turn the specu-lum horizontal, and tilt toward the rectum Slowly open the speculumblades, allowing the cervix to come into view Secure the speculum in anopen position that allows the clinician to obtain any needed specimens Obtaining the Papanicolaou smear consist of acquiring endocervicalcells by placing the longer end of the wood scraper in the cervical os.Press and turn 360 degrees (a full circle) This should include the trans-formation zone at the squamous-columnar junction Remove and placethe specimen onto a glass slide Next, obtain ectocervical cells by plac-ing the endocervical brush inside of the cervical os Roll it between yourfingers (thumb and index) 360 degrees Remove the brush, and roll thebrush against the glass slide You may use the previous slide that con-tains the endocervix specimen or another glass slide Place the slide into

fin-an alcohol solution or apply the special fixative Alternatively, theprovider may obtain a liquid-based cytology by placing the specimensdirectly into preservative

Inspect the vagina on withdrawal of the speculum slowly makingnote of color, inflammation, discharge, or ulcers As the speculum clearsthe cervix, release the lock on the speculum, close and withdraw, revers-ing the insertion process

Trang 15

742 Chapter 21: The Gynecology System and the Adolescent

your fingers to avoid contamination From a standing position, duce one finger and then, if able, the second finger into the vagina.Abduct the thumb, and with the other hand press downward on thelower abdomen Palpate the cervix, uterus, and either side of the fal-lopian tube and ovary Make note of size, shape, consistency, tender-ness, or enlargement of these structures

intro-Rectovaginal Examination

Withdraw your fingers Lubricate your glove again Some clinicians ommend changing gloves or have double gloved initially and removethe contaminated glove prior to performing the rectal component.Inform the patient that the procedure may feel uncomfortable Rein-troduce your index finger in the vagina and the middle finger into therectum Repeat the maneuver from the bimanual examination This com-ponent should not be neglected, especially in the evaluation of abdomi-nal pain or in assessing a retrodisplaced uterus This maneuver alsoallows palpation of the uterosacral ligaments, cul-de-sac, and the adnexa

rec-Synthesizing a Diagnosis

TABLE 21–7 lists the clinical high points of adolescent gynecologicdiagnosis

Laboratory and Imaging

TABLE 21–7 also includes laboratory and imaging aids in the right-handcolumn

When to Refer

TABLE 21–8 lists indications for specialist referral

Trang 16

TABLE 21–7 Gynecologic Disorders of Adolescent Females

absence of smell sense suggests Kallmannsyndrome; visual field deficits suggests brain tumor

Physiologic, imperforatehymen, Mayer-Rokitansky-Kuster-Hauser (MRKH)syndrome, Turnersyndrome (45,XO andmosaicism), chronicillness, hypothalamic:

stress, eatingdisorders, exercise,depression; androgeninsensitivity

syndrome (46, XY);

Swyer syndrome;

others (see text)

Serum gonodotropins(FSH, LH), prolactin,TSH; Pelvic ultrasoundMRI

Head CT/MRI Renal ultrasound/otherimaging studiesKaryotypeLaparoscopy

(Continued)

Trang 17

TABLE 21–7 Gynecologic Disorders of Adolescent Females (Continued)

Pelvic pain during normal ovulatorymenstruation; no underlying pelvic pathology;

may also see gastrointestinal symptoms,headache, myalgia, sweating

May be seen at menarche or 3+ yearspostmenarche

Pregnancy, lactation,stress, eatingdisorders, chronicillness, exercise-induced,prolactinoma(headaches, visualfield deficits,galactorrhea), PCOS(polycystic ovarysyndrome) (see text)

Physiologic

Endometriosis, PID,reproductive tractanomalies, pelvicadhesions, cervicalstenosis, ovarianmasses, pelviccongestion syndrome;

rule out urinary tract

or gastrointestinalcauses

Pregnancy test (β-hCG),progesteronechallenge, serumestrogen, FSH, LH;bone mineraldensitometry; serumprolactin; thyroidscreen; head CT

Laparoscopy, STDscreen, pelvicultrasound, MRI

Trang 18

Menstrual calendar useful to get accurate history

of menstrual pattern; get sexual activity history;

establish presence/absence of ovulation: basal

body temperature charts, serum progesterone,urinary LH and possibly endometrial biopsy;

rule out an STD; virilization evaluationnecessary if hirsutism present (See PCOS)

Pain with history of secondary amenorrhea, oftenwith vaginal bleeding

Presentation in adolescence not the same as inadults; may have acyclic pain, abnormal uterinebleeding, GI symptomatology

Pain associated with ovulation in the middle of amenstrual cycle; may last 1 to 3 days and bemild to severe

Anovulatory bleeding,pregnancy, ectopicpregnancy,coagulation disorders(such as von

Willebrand disease,others), anatomiclesions, endometrialpathology; cervicitis

or cervical dysplasia;

PID, ovarian cysts,polycystic ovarysyndrome, severestress, rapid or severeweight gain or loss,drug abuse See DUB differential

See secondarydysmenorrhea

See secondarydysmenorrhea

CBC, platelets, β-hCG,Pap smear, PT, aPTT,PFA, fibrinogen othercoagulation disordersscreening; thyroidscreen; STD screen;ultrasound(transvaginal; pelvic),MRI; hysteroscopy

β-hCG; pelvic ultrasoundLaparoscopy, laparotomy

Menstrual calendar

(Continued)

Trang 19

TABLE 21–7 Gynecologic Disorders of Adolescent Females (Continued)

precipitated by N gonorrhoeae, C trachomatis, others (Gardnerella vaginalis, Mycoplasma hominis, Ureaplasma urealyticum, H influenzae,coliforms,

cytomegalovirus, Peptostreptococcus, and other

anaerobes); can involve various combinations ofendometritis, salpingitis, tuboovarian abscess,and pelvic peritonitis; complications includeinfertility, chronic pelvic pain, ectopicpregnancy

Ovarian cysts, ovariantumors (benign,malignant), polycysticovary syndrome,ectopic pregnancy,tuboovarian massEctopic pregnancy,appendicitis,pyelonephritis,ovarian cyst, septicabortion, others

Pregnancy test, pelvicultrasound; screen fortumor markers: alpha-fetoprotein, estrogen,progesterone,testosterone, LDH.Nonspecific:

WBCs on saline prep;elevated ESR; elevatedCRP; lab evidence of

N gonorrhoeae or C trachomatis; specific

criteria: positivebiopsy, endometrium,showing endometritis;evidence of PID onlaparoscopy; ultrasound

or MRI showing thatfallopian tubes arethick and filled withfluid; may be free fluid

in the pelvis or atuboovarian complex

Trang 20

Other causes ofhyperandrogenism:

HAIR-AN syndrome;

congenital adrenalhyperplasia(11β-hydroxylase,21-hydroxylase,3β-hydroxysteroiddehydrogenasedeficiency); Cushingdisease, ovarianhyperthecosis,hyperprolactinemia;

ovarian or adrenaltumor; mixed gonadaldysgenesis

(45,X/46,XY; gonadaldysgensis withvirilization; truehermaphroditism

LH, FSH, T4, prolactin,testosterone (total andfree), insulin level,lipid profile,dehydroepiandrost-erone sulfate (DHEAS),17-

hydroxyprogesterone,24-hour urine for freecortisol,

dexamethasonesuppression test, pelvicultrasound

(Continued)

Trang 21

TABLE 21–7 Gynecologic Disorders of Adolescent Females (Continued)

DSM-IV (2000) criteriafor PMDD

Abbreviations: CBC = complete blood count; Pap = Papanicolaou smear; STD = sexually transmitted disease; MRI = magnetic resonance imaging; GI =

gastroin-testinal; ESR = erythrocyte sedimentation rate; F = Fahrenheit; C = Centigrade; HAIR-AN = hyperandrogenism, hirsutism, insulin resistance, acanthosis nigricans;

DSM-IV = Diagnostic Statistical Manual, 4th Edition (American Psychiatric Association); PT = prothrombin time; aPTT = activated partial thromboplastin time;

PFA = platelet function analysis; LDH = lactate dehydrogenase.

Trang 22

When to Refer 749

TABLE 21–8 Conditions to Refer

STDs that are Non-STD causes of Complex gynecologic unusual or vaginal discharge disorders:

difficult to difficult to manage:

manage:

HIV/AIDS Foreign-body vaginitis Anatomic causes of

with foreign bodies amenorrhea (e.g., not easily removed MRKH syndrome)Syphilis Contact vaginitis that Androgen insensitivity

is severe or resistant syndrome

to managementChancroid Chronic vulvovaginal Polycystic ovary

inguinale

hyperplasia,Cushing)

to cervical craniopharyngioma)neoplasia

potential)

complications (e.g., ectopic pregnancy)Adrenal and ovarian masses

Endometriosis and othercauses of secondary dysmenorrheaGalactorrheaOthers—depending on the clinician’s expertise—as for example, pelvic inflammatory disease (PID)

Trang 23

This page intentionally left blank

Trang 24

Laboratory Testing Overview

22 Chapter

Vinay N Reddy

Laboratory and other diagnostic studies in pediatrics pose several lems not common in adult medicine One major problem is that chil-dren do not enjoy needle pokes or restraints for imaging procedures andoften express their disapproval emphatically Children also have smallerreserves to call on for laboratory samples; the extreme example of this

prob-is in premature infants in intensive care, in whom the most commoncause of anemia is phlebotomy for laboratory studies For these reasons,pediatricians tend to be more parsimonious when ordering diagnostictests than their adult-practice counterparts—as well we should be

Choosing a Test

Before ordering a particular test

1 Determine what action you will take if the test is positive

2 Then determine what action you will take if the test is negative

3 If a = b, do not order the test: it will be a waste of time, money, and

patient goodwill

There are occasions when a = b for the direct care of the patient but

not for other purposes, such as public health; in such cases, the test stillmay be worth obtaining

Sensitivity and Specificity

A clinician also must know how well a test meets its intended purpose

This can be quantified by the test’s sensitivity (the probability that the

test is positive in a patient known to have the condition being tested for)

and specificity (the probability that the test is negative in a patient known not to have the condition) Probability, or the chance of a particular out-

come, is expressed as a number between 0 and 1 (or 0 percent and 100percent), where zero probability represents impossibility and a proba-bility of one represents certainty Probabilities also can be expressed asodds, or the ratio of two probabilities: The odds of throwing a fair dieand having it show a one are the ratio of the probability that a one will

Copyright © 2008 by The McGraw-Hill Companies, Inc Click here for terms of use

Trang 25

752 Chapter 22: Laboratory Testing Overview

appear (1/6) to the probability that a 2, 3, 4, 5, or 6 will appear (5/6),

or 1:5 More generally,

Odds = probability ÷ (1 – probability) and

Probability = odds ÷ (1 + odds)

To determine sensitivity and specificity experimentally, one must

have a gold standard (a test that is assumed to be perfect); many gold

standard tests are more expensive and time-consuming than their so-perfect counterparts and are obtained only to confirm the results ofinitial tests or for research purposes (such as determining the sensitiv-ity and specificity of new tests)

not-An example is the rapid test for group A streptococcal antigen inthroat swabs: In a comparison between a hypothetical rapid test andmultiple cultures from the same patient (the gold standard) for 1000 dif-ferent patients, the following results were obtained:

One of Multiple All Cultures

Cultures Positive Negative Total

positive

negative

Total 900 (a + c) 100 (b + d) 1000 (a + b + c + d)

This is an example of a 2 × 2 table, on which all our definitions will

be based The rows represent the results of the test we are analyzing,whereas the columns represent “truth” (as determined by the gold stan-dard) Using the letters for each of the values in the table,

Sensitivity = a/(a + c) (in our example, 720/900, or 80 percent)

Specificity = d/(b + d) (in our example, 95/100, or 95 percent)Another way to express sensitivity and specificity is in terms of theerror rates:

False-positive rate (or false-alarm rate) = b/(b + d) = 1 – specificity,

or 5 percent

False-negative rate (or miss rate) = c/(a + c) = 1 – sensitivity,

or 20 percent*

*The terms false alarm and miss were coined by the British Royal Air Force, which

developed “receiver-operator characteristic curves” to measure the accuracy of radar receivers and their operators during the Battle of Britain in World War II Specificity

was expressed as the probability of a false alarm—a British fighter sent aloft to pursue

a German bomber that wasn’t really there—whereas sensitivity was expressed as the

probability of a German bomber being missed long enough to drop a bomb on London The terms false alarm and miss are still used in the literature of statistical detection theory

and are occasionally seen in descriptions of diagnostic tests.

Trang 26

Sensitivity and Specificity 753

A negative result from a test with high sensitivity is good evidence thatthe patient does not have the condition being tested for, whereas a nega-tive result from a low-sensitivity test should be confirmed by other means.Similarly, a positive result from a high-specificity test may be taken at facevalue, whereas confirmation is needed for a negative low-specificity test.The rapid group A streptococcal antigen test described earlier has fairlyhigh sensitivity and a very high specificity Therefore, a positive test is suf-ficient evidence to treat for streptococcal pharyngitis, but a negative test isinsufficient proof that the patient is streptococcus free Many availablerapid group A streptococcal antigen tests have similar sensitivity and speci-ficity, so pediatricians routinely obtain a throat culture if the rapid test isnegative, depending on how strongly they suspect streptococcal infection

A perfect test has 100 percent sensitivity and 100 percent specificityand thus has positive and negative predictive values of 100 percent aswell Unfortunately, there are few, if any, perfect tests Most physiologicmeasurements are continuous in nature, and their values follow a nor-mal distribution (the bell curve); in many cases, the values are bimodal—two bell curves overlapping, one corresponding to the presence of thecondition being tested for and one corresponding to its absence as illus-trated in FIGURE 22–1 This overlap region, where the true result andthe result of the chosen test may not correspond, is what makes a testimperfect To decide whether a test is positive or negative, the clinician

or the laboratory must choose a threshold value to separate positive results

from negative results With some tests, the threshold value between apositive result and a negative result can be selected by the ordering clin-ician; this allows the clinician some measure of control over the sensi-tivity and specificity, but not over both individually The relationshipamong threshold value, sensitivity, and specificity can be described by

a receiver-operator characteristic (ROC) curve (see preceding footnote),

which plots sensitivity against specificity for varying threshold values.When selecting a threshold value, the ideal goal is to maximize bothsensitivity and specificity Practically, increasing the sensitivity usuallylowers the specificity (and vice versa), and the goal is to maximize sen-sitivity for a given specificity (or vice versa) Any test can have 100 per-cent sensitivity: just say that the test is positive regardless of the actualresults—which, for all but gold standard tests, implies zero specificityand makes the test worthless In the ROC curve, shown in FIGURE 22–2,sensitivity is plotted against specificity for differing threshold values

(The x axis is actually 1 minus the specificity, or the probability of a false

alarm.) The upper-right corner represents 100 percent sensitivity and zerospecificity, whereas the lower-left corner is the point of zero sensitivityand 100 percent specificity A perfect test has an ROC curve that passesthrough a point in the upper-left corner of the graph that corresponds to

100 percent sensitivity and 100 percent specificity A real test’s ROC curve

is more likely to rise fairly steeply initially and then curve near the “point

of perfection” and continue with little further rise until it reaches theupper-right corner The diagonal straight line is the ROC curve of a wildguess, which is the worst possible performance for any test Several dif-ferent tests for a particular condition can be compared by plotting theirROC curves on the same graph; the test whose ROC curve passes closest

to the “point of perfection” will have the best discrimination

Trang 27

Predictive Values and Likelihood

Ratios

Strength of suspicion is also important in choosing a test The predictive value of a test (the probability of the test being positive/negative given

that the patient does/does not have the condition) depends not only

on the sensitivity and specificity but also on the prevalence (the ity of the patient having the condition) The positive predictive value (PPV)

probabil-of a given test, which equals a/(a + b) using our preceding notation, rises with increasing prevalence, whereas the negative predictive value (NPV), or d/ (c + d), falls with increasing prevalence [The prevalence in the popula-

tion in which the initial tests were performed is the proportion of test

sub-jects who were positive according to the gold standard test, or (a + c)/ (a + b + c + d).] Note that the sensitivity and specificity of a test depend only on the quality of the test, not on the prevalence of the condition.

FIGURE 22–1 These two bell curves represent the possible results of a tic test that yields a single number as a result The curve on the right shows pos- sible test values for a known (by a gold-standard test) positive result, while the curve on the left shows possible test values for a known negative result d is the dif- ference between the mean values for positives and negatives; the variance is the same for the two results For a particular threshold value P D (the % area under the positive-result curve to the right of the threshold) is the probability of detection

diagnos-of a positive result, while P FA (the % area under the negative-result curve to the right of the threshold) is the probability of a false-negative result The % area under the positive-result curve to the left of the threshold is the probability of missing a positive result Increasing the threshold will decrease P FA and will also decrease

P D , while decreasing the threshold increases P D and also increases P FA

0.050.10.150.20.250.3

Trang 28

For the purpose of determining predictive values, the prevalencedepends on the clinical level of suspicion, which, in turn, depends on his-torical information and physical findings, as well as on population preva-lence Returning to our hypothetical group A streptococcal antigen test,

PPV = a/(a + b) = 720/725 = 99.3 percent

NPV = d/(c + d) = 95/275 = 34.5 percent

However, the prevalence of group A streptococcal infection in this

“population” is 900/1000 = 90 percent This is (hopefully!) not the ulation prevalence of group A streptococcal infection but may very well

pop-be the prevalence of such infection in those patients on whom the rapid strep test is obtained, more properly termed the pretest probability (the

probability before the test is performed that the patient has cal infection) If the prevalence of group A streptococcal infection is actu-ally 10 percent, the preceding table will look like this:

FIGURE 22–2 Receiver-Operator Characteristics as a Function of the Difference d Between the Means for Negative and Positive Results.

One of Multiple All Cultures

Cultures Positive Negative Total

Trang 29

For this “population,” PPV is 80/(80 + 45) = 80/125 = 64 percent,and NPV is 855/(20 + 855) = 855/875 = 97.7 percent Similar calcula-tions for different prevalence rates, with 95 percent sensitivity and

80 percent specificity, are shown below in table form and graphically

in FIGURE 22–3

Screening tests used for large populations, such as occult blood tion in stool as a test for intestinal cancer, need to have high positivepredictive values despite low prevalence This requires sacrificing highspecificity in favor of high sensitivity and using confirmatory tests withhigh specificity to weed out false positives from the initial screening test

detec-Another way to express predictive value is the likelihood ratio (LR) The positive likelihood ratio (LR+) of a test is the likelihood (probability)

of a positive test in a patient who is known to have the condition sitivity) divided by the likelihood of a positive test in a patient who is

(sen-known not to have the condition (1 – specificity, or the false-alarm rate).

In the terms of our 2 × 2 table above,

LR+ = [a/(a + c)] ÷ [b/(b + d)]

FIGURE 22–3 Predictive Value as a Function of Prevalence.

Prevalence (percent) PPV (percent) NPV (percent)

Trang 30

The negative likelihood ratio (LR–) is the likelihood of a negative test

in a patient with the condition (specificity) divided by the likelihood of

a negative test in a patient without the condition (1 – sensitivity, or themiss rate):

Using the likelihood ratios, we can calculate a posttest probability: the

probability that the patient has the condition given the result (positive,

negative, or a particular value) of the test The posttest probability Ppost

is calculated by converting the pretest probability Ppreto pretest odds,multiplying by the likelihood ratio (LR+ for a positive result, LR– for anegative result), and converting the resulting posttest odds back to aprobability:

Ppost = Opost/(1 + Opost) where Opost= (LR × Ppre) ÷ (1 – Ppre)Computing the likelihood ratios and posttest probabilities of diseasefor the tests used in diagnosis allows the physician to quantify theclinical value of those tests and to select tests that will yield the mostinformation at the least cost Subconsciously, most of us do exactlythis when selecting tests to be performed, but it is useful—at timeseven for experienced physicians and certainly for those in training—

to determine explicitly which tests are the best choice for particularclinical situations

Screening and Case-Finding Tests

Computing likelihood ratios is most important in diagnostic testing.Once a patient is diagnosed, most tests thereafter will be ordered to mon-itor the progress of the disease and the patient’s response to therapy.Another important reason for testing is to detect diseases before clini-cal signs develop, whether in the general population or in segments ofthe population at higher-than-average risk for particular diseases Suchtests may reveal patients with higher risk than the general populationfor a particular disease: One example of a screening test is the measure-ment of serum bilirubin at 24 hours of age to detect hyperbilirubinemia

Trang 31

early enough that treatment will prevent kernicterus Testing also may

be desirable for case finding—to detect diseases in their early stages in

specific populations at risk An example is the tuberculin skin test forpatients who have been exposed to tuberculosis (The tuberculin skintest was performed as a screening test for the school-age population asrecently as the 1970s Its use is now limited to case finding in patients

at risk.) Some tests may be used for multiple purposes Urine glucoseand ketone tests may be used to screen the general pediatric populationfor diabetes mellitus, to detect cases of diabetes in high-risk populations’such as obese children, to confirm a diagnosis of diabetes, or to moni-tor the efficacy of therapy in known diabetics

Screening tests, because they are used in large populations, should

be inexpensive, safe, and easy to perform A high-cost/high-risk ing test for a particularly common and deadly disease may be accept-able in some populations and circumstances For example, both thetuberculin skin test and the chest radiograph will detect active pul-monary tuberculosis and can be used for screening However, the chestradiograph will not detect early infection or extrapulmonary tuberculo-sis, requires expensive and cumbersome equipment, and requires expos-ing each screened patient to ionizing radiation The tuberculin skin test

screen-is inexpensive, easy to perform, and cannot be used only in patientsallergic to the purified protein derivative preparation (very rare) or whohave had tuberculosis in the past (not as rare, but a small segment ofthe general population) It is therefore better to use the skin test as theprimary screening tool for tuberculosis, reserving chest radiographs forthose with positive skin tests, a history of tuberculosis, or allergy to thepurified protein derivative

Trang 33

760 Appendix

Trang 34

Appendix 761

Trang 35

762 Appendix

Trang 36

Appendix 763

Trang 37

764 Appendix

Trang 38

Appendix 765

Trang 39

766 Appendix

Trang 40

Appendix 767

Ngày đăng: 11/08/2014, 11:22