443' THE REHYDRATION OF CHILDREN By K. R. LLEWELLIN, M.B., B.S., M.R.C.P. Senior Medical Registrar, Westminster Children's Hospital Rehydration literally implies the restoration of water alone. In simple depletion this is indeed all that is necessary, but in most cases of ' de- hydration ' in childhood there also occur excessive losses of substances other than water. In this article the factors regulating the loss and restora- tion of these substances will be considered with a brief reference to the associated problems of acid- base balance and calorie intake. Much of the account will centre on the problem of rehydration in gastro-enteritis. Finally, a short description will be given of some of the technical procedures to be used. Water Depletion The fluid requirements of early infancy can be calculated with sufficient accuracy on the basis of 21 oz. per lb. of expected body weight per day up to a weight of 15 lb. The expected weight of the infant can be computed by reckoning an ounce per day gain from the tenth to ninetieth day, and then a gain of i lb. per month till the end of the first year. The infant between 15 afd' 19 lb. re- quires 24 oz. per lb. daily. In later childhood the fluid requirements are proportionately less, being 2 oz. per lb. at i year; 1.5 oz. per lb. at 4 years; and i oz. per lb. at 10 years. At these ages ' expected weight' has to be calculated from actual weight, allowing for de- hydration. In addition to maintenance fluid re- quirements a dehydrated child needs a 'restora- tion quota.' Where the amount of recent weight loss is known this additional need can be readily estimated, otherwise it has to be based on clinical assessment. The facies and skin turgor are the most useful signs, the state of the mouth and fontanelle having to be interpreted with caution; hyperpnoea may dry the former and meningitis fill the latter. In infants dehydration suggests a fluid loss equivalent to 4 per cent. of the original body weight; severe dehydration 6 per cent.; and very severe dehydration 8 per cent. The restoration quota should be administered within four hours: Example: A severely dehydrated, three-month- old baby weighs 9 lb. Its birth weight was 7 lb' but its weight immediately preceding the illness is unknown. As the dehydration is severe the pre- dehydration weight can be estimated as 9x Ioo 9 x = o1 lb. approx. 92 The restoration quota is therefore I6 oz., or 450 ccs. The infant's expected gain from birth = 80 oz. (i.e. 90o-o), giving an expected weight of 12 lb. Maintenance intake is therefore z2 x 24 oz. 30 oz., or 830 ccs. This baby should therefore be given for four houis fluid at the rate of 4 oz. or I o ccs. per hour, i.e. about 30 drops per minute if by drip. Thereafter maintenance fluid is given at i4 oz. (35 ccs). per hour (approx. 0o drops per minute). Examples of pure water depletion can be found in cases where the child is taking an in- adequate feed, not associated with diarrhoea or vomiting. The water required in a particular case can be estimated from the above data. In regard to the administration of the restoration quota remember: (a) That the quantity is related to both the size of the baby and the severity of dehydration-if in the example above, dehydration had been mod- erate 8 oz. would have sufficed in the first four hours. (b) Urine output should be increasing rapidly towards the end of that period. In most cases oral feeds of glucose water will prove adequate, the adoption of a 7 x 3 hourly or 0o x 2 hourly schedule being often an advantage. Tube feeding may be necessary or, in cases where the oral-route is inadequate, part or all of thp re- quired fluid may be given by the subcutaneous or intravenous routes. For the intravenous route 5 per cent. glucose is the solution of choice but, unless absorption is expedited by hyaluronidase, sugar solutions should not be administered sub- cutaneously. Sodium, Chlorine and Potassium Depletion and the Acid-Base Balance Sodium is the chief cation and chlorine the by copyright. on April 15, 2020 by guest. Protected http://pmj.bmj.com/ Postgrad Med J: first published as 10.1136/pgmj.27.311.443 on 1 September 1951. Downloaded from
Transcript
443'
THE REHYDRATION OF CHILDRENBy K. R. LLEWELLIN, M.B., B.S., M.R.C.P.
Senior Medical Registrar, Westminster Children's Hospital
Rehydration literally implies the restoration ofwater alone. In simple depletion this is indeedall that is necessary, but in most cases of ' de-hydration ' in childhood there also occur excessivelosses of substances other than water. In thisarticle the factors regulating the loss and restora-tion of these substances will be considered with abrief reference to the associated problems of acid-base balance and calorie intake. Much of theaccount will centre on the problem of rehydrationin gastro-enteritis. Finally, a short descriptionwill be given of some of the technical proceduresto be used.
Water DepletionThe fluid requirements of early infancy can be
calculated with sufficient accuracy on the basis of21 oz. per lb. of expected body weight per day upto a weight of 15 lb. The expected weight of theinfant can be computed by reckoning an ounceper day gain from the tenth to ninetieth day, andthen a gain of i lb. per month till the end of thefirst year. The infant between 15 afd' 19 lb. re-quires 24 oz. per lb. daily.
In later childhood the fluid requirements areproportionately less, being 2 oz. per lb. at i year;1.5 oz. per lb. at 4 years; and i oz. per lb. at 10years. At these ages ' expected weight' has to becalculated from actual weight, allowing for de-hydration. In addition to maintenance fluid re-quirements a dehydrated child needs a 'restora-tion quota.' Where the amount of recent weightloss is known this additional need can be readilyestimated, otherwise it has to be based on clinicalassessment. The facies and skin turgor are themost useful signs, the state of the mouth andfontanelle having to be interpreted with caution;hyperpnoea may dry the former and meningitisfill the latter.
In infants dehydration suggests a fluid lossequivalent to 4 per cent. of the original bodyweight; severe dehydration 6 per cent.; and verysevere dehydration 8 per cent. The restorationquota should be administered within four hours:
Example: A severely dehydrated, three-month-
old baby weighs 9 lb. Its birth weight was 7 lb'but its weight immediately preceding the illness isunknown. As the dehydration is severe the pre-dehydration weight can be estimated as
9x Ioo9x = o1 lb. approx.92
The restoration quota is therefore I6 oz., or450 ccs. The infant's expected gain from birth =80 oz. (i.e. 90o-o), giving an expected weight of12 lb. Maintenance intake is therefore z2 x 24 oz.
30 oz., or 830 ccs. This baby should thereforebe given for four houis fluid at the rate of 4 oz.or Io ccs. per hour, i.e. about 30 drops per minuteif by drip. Thereafter maintenance fluid is givenat i4 oz. (35 ccs). per hour (approx. 0o drops perminute). Examples of pure water depletion canbe found in cases where the child is taking an in-adequate feed, not associated with diarrhoea orvomiting.The water required in a particular case can be
estimated from the above data. In regard to theadministration of the restoration quota remember:
(a) That the quantity is related to both the sizeof the baby and the severity of dehydration-if inthe example above, dehydration had been mod-erate 8 oz. would have sufficed in the first fourhours.
(b) Urine output should be increasing rapidlytowards the end of that period.
In most cases oral feeds of glucose water willprove adequate, the adoption of a 7 x 3 hourly or0o x 2 hourly schedule being often an advantage.Tube feeding may be necessary or, in cases wherethe oral-route is inadequate, part or all of thp re-quired fluid may be given by the subcutaneous orintravenous routes. For the intravenous route5 per cent. glucose is the solution of choice but,unless absorption is expedited by hyaluronidase,sugar solutions should not be administered sub-cutaneously.Sodium, Chlorine and Potassium Depletionand the Acid-Base BalanceSodium is the chief cation and chlorine the
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
chief anion of the extra-cellular fluid. Potassiumis the chief cation of the intracellular fluid.Upon the proper quantitative balance of these
substances depends the normal osmotic equili-brium that exists between the intra-cellular, inter-stitial and intravascular fluids. Disproportionateloss of one or the other of them will, in addition,disturb the acid-base balance. Loss of potassiumhas a profound effect on muscle excitability.Dangerous electrolyte loss results from excessivesweating or excessive loss of alimentary canalsecretion.
(i) If loss of gastric secretion, as in the vomitingof pyloric stenosis, is the dominant feature theprincipal loss is HCI. The tendency to alkalosisthereby produced is compensated, with variablesuccess by two mechanisms:
(a) By diminished respiratory loss of CO2(hypopnoea).
(b) By increased renal excretion of base (de-creased formation and secretion of NH4, ex-cretion of Na2HPO, rather than NaH2PO4, andexcretion of base 'covered' by urinary car-bonic acid).
The net result is a loss of sodium chloride fromthe interstial and intravascular fluids, the con-sequent fall in osmotic pressure determiring thesecondary dehydration and oligaemia of saltdepletion.There is an increase in the alkali reserve. The
normal alkali reserve values in infancy, 45-55vols. % (20-25 m Eq/L), are somewhat lower thanthe adult range. Treatment in almost all casesresolves itself into supplying all the necessary Clin the form of sodium chloride, and allowing thenormal ' balancing action ' of the kidney te restorethe acid-base equilibrium. It is to be rememberedthat in persistent vomiting not due to obstructionat the pylorus the loss of alkaline duodenal juicewill produce a tendency to acidosis. In decidinghow much and what kind of a salt solution toadminister it is well to consider the salt and fluidrequirements separately.
If we take as an example the baby consideredunder water depletion (but assuming that the de-hydration is secondary to vomiting with produc-tion, therefore, of associated salt depletion):Baby aged 3 months with expected weight of
12 lb., an actual weight of 9 lb.; severely dehy-drated so that its calculated recent weight isapproximately io lb.
Expected maintenance fluid intake = approx.30 oz.
Expected maintenance salt intake (see below)I.O gm.
If the fluid intake is given as, say, 4.3 per cent.glucose + I/5 N. Saline the salt requirements willbe well covered-so far as maintenance needs are
concerned. If this child received its maintenancefluid intake in the form of normal saline its dailysalt intake would be approximately 7.0 grammes.Oversaturation with salt is still an all too frequentandfatal blunder.As the mechanism producing the dehydration
in this case is vomiting there will be almost cer-tainly an associated salt deficit. The correctingfluid should therefore be, say, 5 per cent. glucose+ N. Saline. This should be administered onlyduring the phase of rehydration, i.e. 450 ccs. infour hours: approx. 2 grammes of salt.
Will this be sufficient to correct the salt deficit?If, in fact, this has been achieved the urinarychloride concentration will rise despite the in-creased urinary output associated with absorptionof the restoration quota. This urinary chlorideconcentration can be estimated by means of theFantus test. To ten drops of urine add one dropof 20 per cent. potassium dichromate (as indicator)then, drop by drop, 2.9 per cent. silver nitratesolution. The number of drops of the lattersolution required to produce a yellow-red-browncolour change is supposed to be a measure of thenumber of grammes of chloride (as NaCI) perlitre of urine.
Apart from the fact that the test measureschloride and not sodium excretion (and in chronicdiarrhoea this is a severe limitation of its value)experience soon shows that the claims that havebeen made for it, as a quantitatively accurate testare, in fact, over-optimistic. At least such is thecase in infants. It remains a valuable thoughmerely qualitative indication that salt reserves aredepleted and are not rising in the body. Thepresence of one or more ' grammes per litre' infairly dilute urine is an indication for the sub-stitution of I/5 N. Saline + 4.3 per cent. glucosefor a N. Saline + 5 per cent. glucose. Once thephase of severe depletion is passed salt balancewill be maintained by the following daily intakes:
Neonate, 0.25-0.5 gm.; 1-3 months, 0.5-1.0 gm.;3-12 months, 1-2 gm.; then a progressive risethrough childhood to the adult I0-15 gm. Inall cases it is advisable at the outset to obtain bloodfrom either the internal or external jugular veins orfrom the femoral vein, for estimation of the serumchlorides and alkali reserve. If the history ofvomiting is a prolonged one serum proteinsshould be estimated as well. If the baby's con-dition contra-indicates venepuncture, sufficientblood may be obtained from a heel stab for theabove estimations.
Recently it has been pointed out (Danowski,et al., 195 I) that in severe vomiting potassium lossesmay be considerable, and that serum potassiumlevels may not indicate such depletion. Theyquote a case of pyloric stenosis which, on ad-
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
September 1951 LLEWELLIN: The Rehydration of Children 445
mission, had a normal serum potassium, but whichdeveloped a hypokalaemia on the administrationof potassium-free fluids. Balance data suggestedthat during his illness the baby had lost about one-fifth of his estimated total body potassium. Insuch cases the administration of a potassiumsolution similar to Darrow's but without thesodium lactate would be indicated, of course afterthe re-establishment of good renal output.Diarrhoea
In dtarrhoea there is a loss of Na and, in lesserdegree, C1 from the extracellular fluid. There isalso a shift of K from the intracellular to extra-cellular fluid, with partial loss to the alimentarycanal.The absolute loss,of electrolyte from the tissues
and the consequent fall in osmotic pressure bringsabout secondary dehydration and oligaemia in amanner similar to vomiting, but the predominantloss of base brings about a concomitant acidosis.The respiratory and renal mechanisms describedin alkalosis of vomiting are brought into play, butin the reverse direction:
(a) Increased respiratory loss of CO2 (hyperp-noea).
(b) Decreased renal excretion of base (increasedformation of NH4 and excretion of NaH2PO4rather than Na2HPO4).The loss of K is particularly characterized by
circulatory failure and hypotonicity. If NaCI isused as the correcting substance in a case ofsevere and prolonged diarrhoea the preceding dis-proportionate loss of Na will falsify the results ofthe Fantus test in that a high urinary excretion ofchloride may be accompanied by low blood sodiumlevels. Clearly, in such a case, serum sodiumlevels should be determined. Further, the loss ofK which will inevitably occur in such a case willnot be made gool if NaCI alone is used. Par-ticularly as a result of the work of Darrow,potassium solutions have come into use in suchcases. Darrow's solution consists of:-
(Na, 120 m.Eq./L.; K, 35 m.Eq./L.; Cl,105 m.Eq./L.; and lactate, 52 m.Eq./L.)The normal serum potassium, I9-21 mgm. per
cent. (5 m.Eq./L.) may fall to 15 mgm. per cent.(4 m.Eq./L.) or less in severe diarrhoea. If overcorrection occurs danger of heart block exists ataround levels of 35-40 mgm. per cent. (8-Iom.Eq./L.). It is therefore essential:-
(a) Never to administer parenteral potassiumsolutions until a good urinary output has beenestablished.
(b) So to regulate the intake of potassium as toprevent excessive blood levels being reached.
This implies that the 'restoration quota'should be administered as 2 N. saline + 5 percent. glupose. If urinary excretion is well estab-lished by this measure, then the administration ofa potassium solution can be commenced. Darrowrecommends the subcutaneous infusion of hispotassium fluid at the rate of 36 cc. per lb. in thefirst 24 hours, followed by 22 cc. .per lb. on sub-sequent days, so long as parenteral therapy isnecessary. Total fluid requirements, calculated inthe usual way, are met by the additional ad-ministration of 5 per cent. glucose intravenously.As oral feeds are introduced the first fluid used ismade up of one part of potassium solution to twoparts of 5 per cent. glucose. With the introductionof milk feeds, i gm. of KCI is added to each day'sintake until the calorie value of the feed has reached35 cal. per lb. body-weight.Many workers prefer to administer the paren-
teral potassium solution as part of the intravenousinfusion. A convenient plan would be to givefluid in four three-hourly periods, in each of whichperiod the potassium solution is administered in adosage of 9 cc. per lb., alternating with three-hourly periods during which 5 per cent. glucose isadministered in sufficient amount to cover totalrequirements. This on the first day; on sub-sequent days the dose of Darrow's solution in eachperiod would be reduced to 5.5 cc. per lb. Afurther alternative is to administer equivalent dailyquantities of KCI as a continuous drip of lowerconcentration. It is highly desirable, but notessential, that potassium administration should becontrolled by frequent estimations of the serumpotassium. Methods in general use today are tooSlow to be of much value in guiding therapy, buttechniques based on flame photometry promisegreater rapidity, and the same basic technique canbe utilized for the estimation of serum and urinarysodium levels.The acidosis that is so frequently a feature of
diarrhoeic states will, in most patients, respond tothe above therapy. In patients where it is of greatseverity (CO2 combining power of less than9 m.Eq./L. or 2o vol. per cent.), one-sixth molarsodium lactate should be used initially duringadministration of the restoration quota. Given adosage of 27 cc. per lb., this solution will raise theCO2 combining power by approximately 30 vol.per cent.
Calorie Requirements and the Introduction ofOral FeedsThe normal calorie requirements in infancy can
be reckoned on a basis of 50 calories per lb. ofexpected weight during the first four months and
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
45 calories per lb. from the fifth to the twelfthmonth. In older children the normal range ateach age level becomes respectively greater, butapproximate figures may be given as follows: atthree years I,400 calories per day, at five yearsi,6oo calories per day; and at ten years 2,300calories per day.
Five per cent. glucose solution has a calorie valueof 6 calories per oz.; clearly from the point of viewof adequate calorie intake the early establishmentof oral feeding is important; present-day ten-dencies are undoubtedly in that direction. Oralfeeds are safer than parenteral infusions. Anor-exia, vomiting and abdominal distension are theprincipal contra-indications to oral feeding. Thepassage of large, frequent and loose stools isanother, but it would be wrong to await fornormality in stool formation before commencingfeeds-one might wait indefinitely. Just howrapidly feeds can be increased depends on thehistory of the case. An acute onset, with rapidregression of symptoms on the institution of treat-ment, indicates a rapid introduction and increasein the size of feeds. A child, even if quite severelydehydrated prior to the establishment of parenteraltherapy, may be fit to receive half its fluid require-ments orally on the second day, and all on thethird day. On the second day 4.3 per cent.glucose + I/5 N. saline should be the oral fluid,or, if diarrhoea has been a feature of the case, thei :2 potassium-glucose mixture described underpotassium deficit.On the third day ^ strength half-cream National
Dried Milk (i.e. i measure to 4 oz. water) can beused. This has a calorie value of io calories perlb. expected weight. Subsequent feeds are: onthe fourth day half-strength, half-cream NationalDried Milk (20 calories per lb.): on the fifth daythree-quarters strength, half-cream National DriedMilk (30 calories per lb.). In diarrhoeic casesi gm. KCI should be added to the dav's feed untilthis stage. On the sixth day full-strength, half-cream National Dried Milk (40 calories per lb.) isgiven. This last formula should be used for oneweek. If progress is satisfactory sugar can thenbe added, in the proportion of one teaspoonful toevery 5 oz. of feed. This 47 calories per lb. mix-ture should be used for a fortnight, when a gradualreturn can be made to the diet normal for thechild's age: until this stage is reached vitaminsshould be added to the feeds, several proprietarypreparations being quite convenient and satis-factory.When a history of symptoms is more protracted
and their disappearance slower on parenteraltherapy the introduction of oral feeds has to becorrespondingly more cautious. It may be-eightto ten days before the 40 calories per lb. weight
formula is reached and the parenteral admini-strations of protein, and possibly blood, assumesgreat importance. These will be considered in thefollowing sections. In these cases, and especiallyif aureomycin, chloromycetin, or terramycin havebeen used additional Vitamin B complex shouldbe given.Protein AdministrationThe type of case needing early protein intake
has been indicated above. Estimation of serumproteins will be of value in indicating the presenceof severe depletion; oedema may be due to hypo-proteinaemia. Total serum proteins are normallylower in the neonate than in the adult. In thepremature baby the normal serum protein levelis 4-5 gm. per cent., consequently oedema levelsare readily reached. The protein requirements ofearly infancy can be reckoned on a basis of 1.8 gm.per lb. body-weight; at three years i.6 gm. perlb.; in later childhood I.2 gm. per lb. In infantsand small children virtually all this protein shouldbe of animal origin. If the child's condition pre-cludes the oral route then use must be made ofparenteral infusions of plasma or of casein hydre-lysate. Disadvantages attach to each of thesealternatives. Pooled plasma conveys the virus ofhepatitis with sufficient frequency to make thepossibility a worrying one-in debilitated patientsthe disease can be very serious. Irradiation doesnot completely obviate this risk; no filtrationtechnique has yet been evolved. Casein hydro-lysate would appear to provide a perfect alter-native; physically easier to transfuse than plasma,its use is not attended by any risk of hepatitis.However, a tendency to cause venous thrombosis,and the relative ease with which amino-acid ' spill-over' into the urine can occur, are serious objec-tions. My preference is for plasma; when ad-ministered i part to 3 parts of 5 per cent. glucosethe solution contains I gm. of protein in every 24oz. and, in NaCl content, is equivalent to I/6normal saline.
If salt depletion is suspected the diluting fluidshould be 5 per cent. glucose + 1 N. saline, usedonly so long as urinary evidence of salt depletionis forthcoming. In diarrhoeic cases the additionof 0.75 gm. of KCI to every 560 cc. (20 oz.) of theglucose-plasma solution provides a potassium in-take comparable to the first day on the ordinaryDarrow schedule. The addition of o.45 gm. ofKCI to every 560 cc. of solution should suffice foreach subsequent day that the patient receives in-travenous therapy. The usual check on serumpotassium levels should be observed, i.e. dailyestimation while the intravenous potassium intakeis at a high level, where facilities exist for so doing.The gradual introduction of oral feeds permits the
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
September 1951 LLEWELLIN: The Rehydration of Children 447use of half-cream National Dried Milk (I gm. ofprotein per measure of milk powder). This can bereinforced by:
Casilan (protein 26 gm./oz.) and Prosol (protein19 gm./oz.), practically tasteless whole proteinpreparations, or Hepavite (protein 15 gm./oz.),hydrolysate with malted flavour, and Oral Casy-drol (protein 14 gm./oz.), hydrolysate with cheesyflavour.
In considering which preparation to use thegreat variation in price should be remembered.
Indications for Blood TransfusionMany children who suffer dehydration as the
result of an infective process develop a severeanaemia; many' D. and V.' infants have sufferedfrom a preceding nutritional anaemia. The exhi-bition of large doses of iron is often contra-indicated in these children, and, even when u elltolerated, there is often a poor response. Inassessing an infant in this respect it should be re-membered that the normal haemoglobin level be-tween the second and fifth months is of the orderof 70 or 80 per cent.; however an infant may havea haemoglobin not far below these levels, yetbenefit considerably from a small blood trans-fusion.
If a 'one-shot' technique exploiting a scalpvein is adopted then a dosage of o cc. per lb.should be used. If a drip transfusion is used thefollowing formula can be adopted:
1oo H-- - x 40 x W = volume of blood required
100
in cc., where H = haemoglobin in percentageHaldane, W = infant's weight in pounds.With this technique the rate of administration
should not exceed 12 drops per minute or i oz.per hour. Signs of over-transfusion are: varia-tion in pulse rate, especially marked slowing;dyspnoea; distension of neck veins; cyanosis;basal crepitations; enlargement of the liver andperipheral oedema. At the appearance of the firstof these signs the drip should be slowed to sixdrops per minute and the baby carefully observed.If improvement is not immediate the drip shouldbe stopped. The other aspects of blood trans-fusion differ in no way in children as comparedwith adults.
Technical ProceduresTechnique of obtaining bloodfrom heel prick. In
the routine management of the dehydrated childcareful assessment of the clinical features givessufficient guidance-for correct therapy. When acase presents with atypical features, or where re-sponse to treatment is unsatisfactory, determina-tion of the serum chloride, serum sodium and
FIG. i.-Shewing the expression technique for obtainingblood by heel puncture.: note the full relaxationallowed between each period of pressure.
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
uFIG. 2.-The type of tube used for collecting blood
after heel puncture.
serum potassium, and estimation of the alkalireserve and blood urea may be of great importance.Venous blood from the internal jugular or ex-ternal jugular or femoral vein is the most satis-factory for these tests, but if, for any reason,difficulty is experienced in obtaining blood, aheel prick may be satisfactoty. The baby's heelshould be first warmed by immersion in warmwater for two minutes. It should then be rapidlydried and a thin smear of sterile vaseline appliedto the heel (this latter measure aids the collectionof the blood as discrete drops). The heel shouldbe firmly pricked with a Hagedorn's needle andthe expression technique illustrated in Fig. ishould be carried out. The importance of com-pletely relaxing between each compression isshown by the illustration. The blood should becollected into one or other of the tubes shown inFig. 2.
Tubefeeding. In cases where inability to swallowis the principal bar to the maintenance of hydra-tion, tube feeding may be of the greatest value.It is not without danger; it should not be orderedby a medical officer wHo has no practical experienceof the procedure, nor, should a nursing staff,similarly ignorant, be expected to carry it out.It undoubtedly has its greatest sphere of useful-ness in the immature neonate, but some workershave used it extensively, often by drip-feed tech-nique, in older children. Dangers of the methodlie in misdirection of the tube, over-rapid disten-
sion of the stomach, regurgitation and inhalationof feed, and injury to mucosa. The essentialapparatus is indicated in Fig. 3. Nasal or per-oral intubation should be decided on the followingbasis:
(a) The nasal route is absolutely contra-indicated in the neonatal group. In such cases aNo. 7 (French) oesophageal tube should be passedorally.
(b) In older children the nasal route is oftenpreferable as the tube is better tolerated in thisposition. It should not be used for repeated in-tubation. A No. 5 (French) catheter is suitablein most cases.
(c) In older children, where the nasal route forany reason is contra-indicated, a No. 8 oeso-phageal tube can be used.The apparatus, as illustrated, should be
assembled and washed through with sterile water.With the child in a sitting position the catheter oroesophageal tube should be passed rapidly. Fornasal intubation a trace of lubricant should beapplied to the tube; one nares may prove easierthan the other. Force is unnecessary and in-jurious. For an intra-gastric feed the tube shouldbe passed an inch further than the distance fromthe mouth to the ensiform cartilage. An escape ofgas will indicate that the tube is in the stomach.A somewhat shorter length should be passed foran oesophageal feed. Coughing or change ofcolour on the part of the baby is an indication for
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
September 1951 LLEWELLIN: The Rehydration of Children 449
~lY ~! :X r-Ni ,
FIG. 4.-Instruments for intravenous infusions in infants and young children. B', B', Bateman'sneedle: C, polythene tubing and adaptor: D, modified lumbar puncture needle.
the immediate withdrawal of the tube. As afurther precaution a small quantity of sterile watershould be passed before the feed.
Oesophageal feeding prevents the over-rapiddistension of the stomach and is especially in-dicated in the more feeble or immature neonate.The last portion of the feed is' followed down ' bya small quantity of sterile water; the tube is thenrapidly withdrawn.
Subcutaneous InfusionsThis route can be relied upon to contribute only
a part of the child's total daily fluid requirements.Using hyaluronidase total needs may be met, butnot with certainty. Subcutaneous infusions areuseful in two particular roles:-
(a) As a means of satisfying potassium require-ments with safety-as in the Darrow technique.
(b) It may supplement oral feeds, either in themild case of dehydration or subsequent to the useof the intravenous route in more severe cases.
The axillae, abdomen or thighs can be used.A Y-shaped connection with double needles is anadvantage. If hyaluronidase is used I mgm.should be injected into the drip tubing immediatelyproximal to the needle. In slowly absorbed sub-cutaneous infusions the use of glucose is attendedby a grave risk of cellulitis.
Intravenous InfusionsThe paediatrician has a choice of an astonishing
variety of sites and techniques; however, thebasic technique of which mastery is essential foranyone dealing with sick children is the cut-down intravenous drip using a small, rigidcannula or polythene tubing. The most reliablevein is the internal saphenous vein at the internalmalleolus. As far down as the needle adaptor theinfusion apparatus is the standard M.R.C. in-fusion set. The large needle or cannula is re-placed either by a Bateman needle or by anadapted fine lumbar puncture needle (Fig. 4d) orby polythene tubing (Fig. 4).
D
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
September 1951 LLEWELLIN: The Rehydration of Childrcn 451
The Bateman 'needle' consists of an outercannula (B") which is introduced into the vein,and an inner cannula (B') which is attached to thedrip set. Clearly such an arrangement facilitatesthe changing of defective or obstructed dripapparatus. It is a widespread practice nowadays touse only the outer Bateman, thus minimizing thedegree of obstruction to the infusions of, es-pecially, blood and plasma. A lumbar punctureneedle of bore intermediate between the outer andinner Bateman cannulae, suitably shortened andwith the point rounded off, is very useful. Becauseof the ease with which it adapts itself to the courseof the vein, and because a sufficient length can bepassed to make rigid splinting of the leg un-necessary, polythene tubing (sizes i or 2) has muchto commend it. It is claimed that its use isattended by a higher incidence of thrombo-phlebitis, and that the tubing tends to 'fur up'rather easily. In my experience it has provedmore convenient and less attended by undesirablesequelae than the rigid cannulae. The fact thatthe tip of the polythene tubing usually lies in thelumen of a bigger vein than would the tip of arigid cannula, appears to make for easier running.
It is important to carry out the assembly of aninfusion in an unhurried fashion. The less fre-quently the operation is performed in a unit thegreater are the chances of leaks and misfits in thedrip apparatus and defects in the cutting downset. It should not be necessary to ' assemble ' therequisite instruments; full cutting-down setsshould be available already sterilized. The splint-ing of the child's limb should be done by thedoctor himself, with the ward sister's supervision,at least until he is familiar with the possible snags.Chloral sedation is often valuable; a restless babyof three months will need 2 gr.
It is important that the child should be keptadequately warm throughout the procedure-an' Anglepoise ' lamp (Fig. 6) serves usefully both towarm and light the leg. Elsewhere well-protectedhot water bottles may be necessary. In bloodtransfusion, especially, overheating must beavoided. If oxygen therapy is deemed necessarythe tent should not be discontinued for the cut-down; a re-arrangement of the tent should permitaccess to arm or leg. A well-padded splint whichprojects for some distance beyond the limb shouldbe used. The thigh should be slightly abductedin full external rotation; the leg should be slightlyflexed; the foot plantar-flexed and inverted. Whilethe thigh and leg are held by the nurse the footshould be firmly strapped in the above position.The leg and thigh can be immobilized either by awell-padded bandage or preferably a series ofencircling non-elastic adhesive tape strips. If thelatter are used it is important that the tape overlap
should rest on underlying tape and not on skin;if, as rehydration proceeds, it becomes necessary toslacken these bands, undue pulling upon the skinis thus avoided. It should be possible to removethe baby from the cot for the purpose of feeding,etc., without unduly disturbing the drip.
Local anaesthesia can be attained by infiltrationof the surrounding area with i per cent. procaine.The drip apparatus should then be assembled,connected with the infusion bottle and checked tosee that it is running normally. When the bottlehas been suspended, a piece of adhesive tape,calibrated so as to measure the contents of thebottle in ounces, is fixed to the side of the bottle.The initial fluid level should be noted.The patency of the cannula should be assured;
if polythene tubing is to be used it should be re-moved from cetavlon and rinsed with sterile watercarefully cleansing both lumen and exterior. Itshould next be threaded into and fixed in the,adaptor shown (Fig. 4c). Next make certain thatthe catgut ligatures and aneurysm needle arehandy and that a syringe loaded with normal salineis available. Lastly, before making the incision,see that the light is in a good position and that oneis in a comfortable working position.A transverse incision shown is made through
the full thickness of the skin. From then on pro-ceed by blunt dissection, using an artery forcepswith a light action; a stiff forceps makes for earlyfinger fatigue and clumsiness. Usually, but notinvariably, the internal saphenous vein is fairlyconstant in regard to surface marking; its depthfrom the surface is very variable. The presence ofa thread-like parallel tributary vein lying superficialto the true vein may prove very confusing to the in-experienced. If the vein cannot be found after apatient search, the incision should be enlargedposteriorly. When the vein is located it should bedefined over a distance of perhaps I in.; over-enthusiasm in cleaning the vein may precipitatetroublesome spasm and should be avoided. Withthe aid of the aneurysm needle, two strands of cat-gut are passed behind the vein. The distalligature is tied occluding the vein; the free ends ofthe ligature are held in artery forceps. This pro-vides a useful ' handle' by which the vein can becontrolled. The proximal ligature is half tied andagain anchored by an artery forceps. The vein isnow put on tension by gentle traction on the distalligature, and a transverse snick made in it with finescissors. The success of this step, and of the wholeprocedure, depends on the sharpness and precisefunctioning of this instrument (Fig. 4a), whichshould be used for no otherpurpose. The vein shouldbe hemi-sectioned at the first attempt; nibbling pre-cipitates spasm and so lacerates the vessel wallthat cannulization becomes almost impossible.
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
Retrograde blood flow is rarely of sufficientdegree to be an inconvenience; the Batemancannula or polythene tubing is now gently threadedinto the vein. If a rigid cannula is used it should beallowed to ' follow the vein ' even if it means anawkward angle; polythene of course does soanyway, and about 4 in. of this should be passed.Sometimes, but not invariably, blood will flowfrom the end of the cannula. The infusion tubingshould now be connected and the screw clipopened. If the infusion does not run:
(a) Check the apparatus for the presence of anair lock.
(b) Make sure that the limb is not unduly com-pressed.
(c) Try gentle variation of the angle of thecannula.
(d) ' Prove' the vein by gentle injection of nor-mal saline through the cannula by syringe.
(e) Warm the leg.Syringing through a vein is only justifiable at
this stage; while it may be attempted with successin a drip that has stopped after several hours'running, under such circumstances the procedureis too dangerous. Fixation of the cannula and ofthe tubing immediately beyond the limb, is ofgreat importance. Narrow strips of adhesive tapeserve very well. The wound itself should bedusted with flavazole or penicillin-sulphathiazolepowder and covered with sterile gauze and abandage.The rate of drip is often rather variable in the
early stages of a transfusion and needs constantchecking. The number of ounces infused in eighthours is approximately indicated by the number ofdrops per minute; but the calibrated tape is an
essential check on the quantity given. During thephase of restoration the rate may be relativelyhigh-of the order of 20 to 30 drops per minute.Maintenance rates are in the order of o to 15drops per minute.During this early stage oral fluids will probably
consist only of drachm doses; as the oral con-tribution increases the parenteral quota isproportionately reduced. The keeping of a com-prehensive fluid intake and ouput chart is essential.
Observation of this fluid balance is a seriousresponsibility which rests upon the nursing staff;equally important is their immediate reporting ofany untoward change in the baby's condition orany suggestion of blockage in the drip.
If the drip slows down after running for severalhours the possibilities causal of slow startingshould be considered, except that syringingthrough the vein is contra-indicated at this stage.The leg should be re-examined for developingoedema and the dressing loosened if necessary.Any development of an inflammatory reaction
around the drip wound or further up the leg is animperative indication for stopping the infusion.In a small and debilitated baby a suppurativethrombo-phlebitis or cellulitis may proceed veryquietly until the calf is a sack of pus; if the in-fecting organism is Ps. pyocyanea the outlook isespecially grave. The safeguard is meticulousaseptic technique and care, but not meddlesomeobservation of the drip; the value of a chemo-therapeutic 'umbrella' is controversial, but someexperienced operators are convinced of its value.
I am grateful for the help given me by Mr.Levi, Dr. Beautyman and Sister Hammond, andto Dr. Hansell who provided the photographs, andmy wife for the line drawings.
REFERENCES
Technical Procedures
GARROD, BATTEN and THURSFIELD (1947), 'Diseases ofChildren,' Vol. i, Chap. VIII.
Water, Electrolyte, and Arid-Banc Balance(v.s.) Chap. VT
MARRIOTT, H. L. (1950), 'Water and Salt Depletion.'GAMBLE, J. L. (I147). 'Chemical Anatomy, Physiology and
Pathology of Extracellular Fluid.'
Potassium TherapyDARROW, D. C. (I949), Paediatrics, 4, 604-619.DARROW, D. C. (1948), Bull. N.Y. Med., 24, 147-165.DANOWSKI, T. S. (195I), Acta Paediatrica, 40, I98-212.Blood TransfusionWHITBY and BRITTON (i950), ' Disorders of the Blood.'Early Oral FeedingCHUNG, A. W. (1948), J. Paediatrics, 32, I.
by copyright. on A
pril 15, 2020 by guest. Protected
http://pmj.bm
j.com/
Postgrad M
ed J: first published as 10.1136/pgmj.27.311.443 on 1 S
